Support NeoGAF

[hemo memo]

The closer we get to launch the faster people are losing their minds.

 

[Zathalus]

MORE PROOFS DF FULL OF CRUD 2022 STEAMDECK SERIES S LIKE EXPERINCE 2024 BACKTRAK ahaha

 

[Redneckerz]

I am getting kind of tired of Gigabowser's illogical posts that are written like a 12 year old and i forecast that Men in Boxes will get a visit soon.

 

[MrJangles]

Don't care who is right or wrong but this 'MetalDave' bum is clearly riding on DF's coat tails for exposure.

 

[Shaki12345]

Don't care who is right or wrong but this 'MetalDave' bum is clearly riding on DF's coat tails for exposure.

He isn't technical at all and pretends he knows technical stuff. He's a retard hunting for clicks and fanboy validation.

 

[Ozriel]

I don't like those kind of youtube channels, but their arguments and proofs are real. The DF logic:

Switch 2 docked = PS4
Switch 2 docked > Steam Deck
Steam Deck = Series S
Switch 2 docked < Series S

But DF has never claimed Steam deck was as powerful as Series S…

 

[simpatico]

Almost every single time he mentions they are working on an Unreal fork. Which is totally going to be the best thing ever and solve everything with the ol "tried and tested" methods. Just ignore the fact he has zero experience or meaningful work to his name.

I don't recall him doing an analysis of FSR4 either, he just dismissed ML upscalers in the Kingdom Come video because of vendor lock in, which is ridiculous.

Does DF have a GH repository full of wildly efficient proof of concepts? I've seen this angle before here on GAF, but no one ever comes with actual receipts to counter any of the points made in the videos. My experience with upscalers warrants broad dismissal too.

 

[Zathalus]

Does DF have a GH repository full of wildly efficient proof of concepts? I've seen this angle before here on GAF, but no one ever comes with actual receipts to counter any of the points made in the videos. My experience with upscalers warrants broad dismissal too.

The difference is DF is not claiming to create a custom Unreal fork that is totally going to fix all of the issues with modern rendering.

 

[GigaBowser]

So, your opinion about this topic is not valid… I don't like that guy either, but he is just rewatching some old DF videos and making comparison with newer ones… Is not even his opinion, DF contradicts himself.

they dont watch he just post DF OWN WORDS

OWN WORDS

I don't understand why so many are negative among enthusiasts. What's the deal? It was 8 years ago since we got new hardware from Nintendo, now it's finally time again. Can we try to be normal and be excited for once?


So, just like any other console then? …which, using your words, occupies the same space as their predecessor, so most of their power lies outside the hardware spectrum, so who cares which outdated PC hardware it resembles the most? Right?

thank you my friends I arm the spirit of gaming

Social Opinion Community  Post in thread 'NEOGAF POSTERS OF THE YEAR AWARD'

Apr 16, 2025

You might be too modest to say so GigaBowser but it's clearly you. Keep up the good work.

• DKehoe

• 
• 
• 

I am getting kind of tired of Gigabowser's illogical posts that are written like a 12 year old and i forecast that Men in Boxes will get a visit soon.

 

[Redneckerz]

And yet you replied, so the meme certainly is ironic.

 

[Filben]

 

[Fess]

I have no problem understanding what is in the heads of people who tend to express themselves here.

It would be nicer to collect all this in one place rather than generate another gotcha thread about something that isn't even particularly surprising or even relevant.

Sometimes I agree, when the warring is too obvious and frequent. But I think during pre-launch times you have to accept threads like this. It only happens once or twice per decade.

 

[simpatico]

The difference is DF is not claiming to create a custom Unreal fork that is totally going to fix all of the issues with modern rendering.

Ok, but if we only focus on the critiques, are there any discrepancies? I haven't watched every vid he's done, but I never heard him talk about any personal projects.

Is there someone doing better video analysis of the same sort?

 

[thicc_girls_are_teh_best]

Why are you saying that?

DF are clearly hypocritical here. "changed opinion" sure, so they have nobody over there doing editorial checks for consistency.

When you spent that long of a video stroking Steam deck to be a series S handheld, i would expect that the stronger handheld coming up is not thrown under the bus as being just PS4, especially when its running games that just won't even run on PS4.

Not sure what your beef with metaldave is, the guy is basically looking at both DF videos and pointing out the hypocrisy. His DLSS comparison with cyberpunk 2077 is legit too

Digital Foundry killed their reputation years ago, but it's nice to see more content creators call them out for their bullshit.

And their analysis have accelerated in degradation since IGN acquired them. What a coincidence.

this only works if you completely ignore context or don't understand basic English.

DF compared the Deck to the Series S because they use the exact same architecture. Zen 2, RDNA2. they showed games running roughly with the same performance but massively dialed back resolution to fit the hardware profile of the Deck.

DF compares the Switch 2 to the PS4 because in raster perfomance they are very similar and the Switch 2 is designed to be docked and used as a home console on a 4K screen.

there's no contradiction here, nothing that was said by DF is factually incorrect.
you either want to misunderstand them on purpose (Like Super Grifter Dave), or you just don't follow what they said.

Personally I don't care about this Dave dude being a grifter or whatever, as that doesn't immediately invalidate potentially valid points brought up, IMO. I can separate those from the grift, if it exists.

But you don't think it's hilarious that DF are comparing two systems with VERY different architectures and feature sets by what one or two performance metrics they might happen to have in common? And somehow think that's a valid comparison?

There's a reason we never did this with older systems like even 360 vs PS3 (or for those who did, shouldn't have), let alone those of gens prior. You simply cannot directly two systems of different architectures directly, or think that pulling a couple of points for comparison where the paper specs are "roughly similar" suddenly makes the comparison valid.

So in short, them comparing Series S and Steam Deck made some sense. Them comparing Switch 2 and PS4 made absolutely no sense.

 

[kevboard]

So in short, them comparing Series S and Steam Deck made some sense. Them comparing Switch 2 and PS4 made absolutely no sense.

comparing 2 home consoles that will share many games between them makes no sense?

they also never compared the hardware, they compare roughly what the quality of the graphical presentation of games will be like. and they will be PS4 like.

Street Fighter 6 is essentially using PS4 settings for example

it once again just clearly seems to be the case that people either purposefully try to misinterpret what DF says or are just actually too stupid to understand context

 

[rodrigolfp]

What in the hell is this pixel soup. With frame gen starting from the 20~30 fps baseline

XesS pixel soup?? The constant 30fps shit show on Switch will be better?? And watch it having the same or more input lag that Deck with frame gen.

 

[Buggy Loop]

comparing 2 home consoles that will share many games between them makes no sense?

So is Steam deck?

they also never compared the hardware, they compare roughly what the quality of the graphical presentation of games will be like. and they will be PS4 like.

So is Steam deck?

Street Fighter 6 is essentialiusing PS4 settings for example

So is Steam deck?

it once again just clearly seems to be the case that people either purposefully try to misinterpret what DF says or are just actually too stupid to understand context

You're giving them a shitload of excuses kevboard, its incredible.

During the course of the years DF has shared a LOT of questionable opinions about the Switch 2 with a lot of "because Nintendo". Unlikely that tensor cores will be able to have DLSS on that portable. Unlikely that it'll have VRR or 120Hz. Would be less powerful than a Steam deck. Don't believe the quality of the matrix demo.

For 3 weeks they declared it would be PS4 target while omitting the narrative that of SF6 or that Cyberpunk 2077 phantom liberty is simply not running on PS4. I/O streaming and more recent architecture, more powerful CPU? Eh who cares. Why try to put some context into what we're seeing? FF7 intergrade with upgraded lighting? Eh, same as PS4 clearly.

RDNA 2 Steam deck? OMG WOW! Portable Series S??

They were OVERLY optimistic for Steam Deck at launch.
They are OVERLY conservative about Switch 2

Both takes hurts their credibility.

Metaldave exposes them as having very different posture. DF realized maybe after 3 years that their claim of portable Series S is crumbling and that's also not a good look on them to be honest, they seem completely incompetent in what they are supposed to do just like I've also showcased how utterly incompetent they are with Switch 2 target with their stupid as fuck emulated 2050 GA10x which is not even at the time of cutdown T234 the same SM as GA10b is with the double FMA tensor cores like GA100 AI chipsets and they made an entire freaking video trying to gauge DLSS performance on a downclocked 2050 when its not even remotely close to AGX Orin. How can these peoples not even bother to open a whitepaper when they're spewing this shit to a large audience?

This is the reference of tech news on youtube for consoles?

They've become a joke. Both Steam deck launch take and now Switch 2 take is not shining a good light on them. You keep trying to say we're too stupid to understand the context, are you for real? I think pretty much everyone here can understand a more modern architecture with better CPU and more VRAM and decompression engine making it run games that devs gave up to make it run on base PS4 makes it beyond PS4 baseline. Yet DF refuses to even acknowledge it. Its not Series S either like I said, but their narrative that its baseline PS4 is completely stupid. If you believe this you're just actually too stupid to understand context.

 

[Buggy Loop]

XesS pixel soup?? The constant 30fps shit show on Switch will be better?? And watch it having the same or more input lag that Deck with frame gen.

I see with my own eyes it looks like shit. Switch 2 footage looks leagues better

DLSS nintendo switch is challenging even transformer model, it seems tailor made for low res like none of the DLSS models on PC do. XeSS is not even in same ballpark.

Completely delusional to think frame gen on steam deck is even a thing when its 20~30 fps in the background. What a crud option. His latency is clearly not reported system latency. Put LDAT on that and it'll be utter shit.

 

[rodrigolfp]

I see with my own eyes it looks like shit. Switch 2 footage looks leagues better

DLSS nintendo switch is challenging even transformer model, it seems tailor made for low res like none of the DLSS models on PC do. XeSS is not even in same ballpark.

Completely delusional to think frame gen on steam deck is even a thing when its 20~30 fps in the background. What a crud option. His latency is clearly not reported system latency. Put LDAT on that and it'll be utter shit.

With your own eyes live or on YT footage??

SW2 handheld footage looks blurry + 30fps limited.

He demonstrates the latency clearly with action.

 

[FalconPunch]

Why am i not surprised that the two most deranged nintendo fans are at it again. It's the usual suspects GigaBowser and Buggy Loop. Thankfully the hullabaloo will end once the units are released and we can get direct feed footage. I'm not the biggest DF fan but, the nintendo cabal trying to discredit df because the switch 2 is weak af is funny.

 

[AZRoboto]

The hate against DF is both weird and stupid-as-shit.

They've become a joke. Both Steam deck launch take and now Switch 2 take is not shining a good light on them. You keep trying to say we're too stupid to understand the context, are you for real?

Yeah, weird and stupid-as-shit.

 

[Buggy Loop]

Yes, some games need more CPU power, while others play better with more GPU power. You can't max out both, but for a lot of Unreal games maxing out the GPU can help. Locking the GPU to 1.4Ghz solves the CPU issues in basically any game the Steam Deck can run, and a 1.4 TF RDNA 2 GPU and a 1.72 TF Ampere GPU will, on average, show similar results. Some games will of course heavily favour each as needed.


The 6700 XT is significantly faster than the 3060. Not the 6600 XT. That was the claim I made. As you pointed out a 6600 XT averaged across multiple games is equivalent to a 3060. The 6600 XT is about ~10.8 TF, the 3060 is ~13.8 TF, and the 6700 XT is about ~12.8 TF. Using actual, real world clock speeds. Nvidia advertised boost clocks are always extremely conservative. As you can see, Ampere is a bit behind because of the dual 32 compute nature of the architecture. The equivalent TF RDNA 2 GPUs in regular, rasterised workloads would be around 20-30 percent faster. Naturally the Nvidia GPUs are equal faster when actually comparing GPUs in the same tier and not just looking at the raw TF number.

You're 2 upper tiers of graphic cards above 3060. Like really?

All your subsequent numbers are wrong TF for 3060.

NVIDIA GeForce RTX 3060 12 GB Specs

NVIDIA GA106, 1777 MHz, 3584 Cores, 112 TMUs, 48 ROPs, 12288 MB GDDR6, 1875 MHz, 192 bit

www.techpowerup.com

Well, yes. Ampere is clearly better for RT and ML workloads, but this specific discussion is mostly focussed on the raster comparison.

And why would devs making impossible ports on Switch ignore its best assets again?

Well, two reasons in this case.
1.) It is an RT game. Ampere>RDNA2 for any RT workload.
2.) 3070ti TF, ~23, 6800 XT, 20.74 TF. 11% advantage for the 3070ti in TF, the 6800 XT still outperforms it by 7%. (As mentioned, Ampere reported clock speeds are very conservative).

I didn't pick RTX setting, its software RT

Since when RDNA 2 not performing well with software RT? Where does nvidia RT hardware even provide better performances with software RT? You understand software RT right?

Some reasons again.

1.) The 7700XT is not 35.2 TF, because VOPD is useless for any games, and not at all the same thing as Amperes dual issue 32 architecture, which is an actual, usable TF number. Naturally, dual issue 32 compute doesn't magically double performance by itself due to a number of other factors, but it is infinitely more useful than VOPD. The actual 7700 XT TF number would be ~18.3 TF.
2.) Same thing as in Star Wars Outlaws for the 3070 Ti vs the 6800 XT, 11% advantage in TF, outperformed by 7% (14% if you use the lows).


3060ti is ~18.3 TF, 6800 is ~16.9. 8% advantage in TF for Nvidia, but yes, AMD does not perform well in this title at all.

Holy shit where you keep pulling these numbers 16.2 TF

NVIDIA GeForce RTX 3060 Ti Specs

NVIDIA GA104, 1665 MHz, 4864 Cores, 152 TMUs, 80 ROPs, 8192 MB GDDR6, 1750 MHz, 256 bit

www.techpowerup.com

1.) RT game again.
2.) 8% TF advantage for Nvidia, still looses by 8%, even in this RT game.

3060 Ti 16.2 TFlops just a tad under the 6800 16.2 TFlops

This is the peoples claiming that we have to halve Ampere TFlops?

Seems RDNA 3 we have to halve TFlops

3060 Ti 18.3 TFlops 8% higher than the 6800 16.9 TFlops (AMD looses by around 4%).
3070 Ti 23 TFlops 11% higher than the 6800 XT's 20.74 TFlops (AMD looses by around 9%)

AMD Radeon RX 6800 Specs

AMD Navi 21, 2105 MHz, 3840 Cores, 240 TMUs, 96 ROPs, 16384 MB GDDR6, 2000 MHz, 256 bit

www.techpowerup.com

NVIDIA GeForce RTX 3060 Ti Specs

NVIDIA GA104, 1665 MHz, 4864 Cores, 152 TMUs, 80 ROPs, 8192 MB GDDR6, 1750 MHz, 256 bit

www.techpowerup.com

NVIDIA GeForce RTX 3070 Ti Specs

NVIDIA GA104, 1770 MHz, 6144 Cores, 192 TMUs, 96 ROPs, 8192 MB GDDR6X, 1188 MHz, 256 bit

www.techpowerup.com

Terrible showing for AMD here of course.

Why? What's the excuse now?

Maybe it was build for PC in mind? Not a slop console port build by AMD → PC?

I can use individual games as well.

With the wrong TF all over the place again

NVIDIA GeForce RTX 3080 Specs

NVIDIA GA102, 1710 MHz, 8704 Cores, 272 TMUs, 96 ROPs, 10240 MB GDDR6X, 1188 MHz, 320 bit

www.techpowerup.com

Or don't use individual games to try and compare performance? There is a reason averages are so useful.

So a dev that would make a game specifically for Switch 2 would use better the GPU with NVapi and nvidia SDKs than PC

Why are peoples making the narrative that you HAVE to turn down TFlops on Ampere architecture when I clearly showed that its game dependent?

Does it happens sometimes? Yes. On PC.

Would it make sense to happen on a closed platform with all the Nvidia tools to work on Ampere? No.

The advantages of a low-level API are well-known, thanks to the PS5.

Thanks to PS5? What is this? Low level APIs have proven themselves way before PS5.

Most 3rd party games do indeed see a bump in comparison to equivalent PC hardware, but the difference is hardly that staggering

Really? Go build a Jaguar mobile CPU + 8GB + Tahiti HD 7800 and come back to us with running Horizon Forbidden west.

outside 1st party titles specifically tailored extensively around the architecture. Which does mean Nintendo 1st party games will look better than anything a PC handheld can do.

Nintendo will never invest huge budgets for that, so no.

I agree, hence why the CPU performance for the Nintendo Switch should be alright.

You'll never catch DF ever saying anything positive of the kind for sure

I went to the Nintendo event in Amsterdam, DLSS is doing a lot of heavy lifting on Cyberpunk for example.

I'd hope so? That's the point of picking Nvidia. You're also playing old demo.

Native pixel counts look extremely low

How would you even detect internal res at a show. The end result is upscaled by AI. DF could not even detect DLSS for 3 weeks.

so thank god for DLSS.

*Specific DLSS model made expressively for Switch 2. Challenging even transformer models.

If it wasn't for DLSS, I think the image quality on Steam Deck and the Switch 2 might be similar.

Switch 2 is reportedly in medium/high settings for Cyberpunk 2077, unlike all the "low" steam deck showcases when they try to make the game run above 20 fps in dogtown.

Although, using Native XeSS is possible on the Steam Deck, which makes it look quite a bit better, but that requires a lot of tweaking to get decent performance in Dog Town.

At low settings and XeSS has never showcased like Nvidia's new DLSS that its a good thing to have at low resolutions.

Switch 2 DLSS wipes the floor with DLSS CNN model. Challenging even transformer model at low res.

 

[Buggy Loop]

Why am i not surprised that the two most deranged nintendo fans are at it again. It's the usual suspects GigaBowser and Buggy Loop. Thankfully the hullabaloo will end once the units are released and we can get direct feed footage. I'm not the biggest DF fan but, the nintendo cabal trying to discredit df because the switch 2 is weak af is funny.

Oh yea, Zathalus this is the guy claiming that you HAVE to halve Ampere TFlops to compare.

So Docked 1.5TFlops and ~700GFlops handheld

 

[Mr Moose]

You're 2 upper tiers of graphic cards above 3060. Like really?

All your subsequent numbers are wrong TF for 3060.

NVIDIA GeForce RTX 3060 12 GB Specs

NVIDIA GA106, 1777 MHz, 3584 Cores, 112 TMUs, 48 ROPs, 12288 MB GDDR6, 1875 MHz, 192 bit

www.techpowerup.com

They aren't. It's not locked to 1777MHz, he's correct it is over 13TF.
Edit: This is my card:

ASUS PHOENIX RTX 3060 V2 Specs

NVIDIA GA106, 1777 MHz, 3584 Cores, 112 TMUs, 48 ROPs, 12288 MB GDDR6, 1875 MHz, 192 bit

www.techpowerup.com

It is over 1900MHz pretty much all the time.

 

[FalconPunch]

Oh yea, Zathalus this is the guy claiming that you HAVE to halve Ampere TFlops to compare.

So Docked 1.5TFlops and ~700GFlops handheld

Live feed of BuggyLoop and Gigabowser rn . Not since the days of MisterXMedia and SenjutsuSage have we seen this level of derangement. Multiple threads created to argue the power of a handheld, discredit df, etc. Very funny stuff.

 

[Lokaum D+]

this thread is something else

 

[Buggy Loop]

With your own eyes live or on YT footage??

SW2 handheld footage looks blurry + 30fps limited.

He demonstrates the latency clearly with action.

Dude, even steam deck players that played cyberpunk 2077 are all saying switch 2 footage looks leagues better. Stop trying to make Steam deck cyberpunk 2077 dogtown a thing. Its not.

Live feed of BuggyLoop and Gigabowser rn . Not since the days of MisterXMedia and SenjutsuSage have we seen this level of derangement. Multiple threads created to argue the power of a handheld, discredit df, etc. Very funny stuff.

Never gonna be the entertainment of seeing how retarded one can be to halve Ampere TFlops

 

[Buggy Loop]

They aren't. It's not locked to 1777MHz, he's correct it is over 13TF.
Edit: This is my card:

ASUS PHOENIX RTX 3060 V2 Specs

NVIDIA GA106, 1777 MHz, 3584 Cores, 112 TMUs, 48 ROPs, 12288 MB GDDR6, 1875 MHz, 192 bit

www.techpowerup.com

It is over 1900MHz pretty much all the time.

If I go out of reference boards for Nvidia but keep TFlops of reference boards of AMD, how does that make any fucking sense.

 

[kevboard]

So is Steam deck?



So is Steam deck?



So is Steam deck?

indeed, which is why they made videos comparing the Deck with the PS4, even basically calling it a 720p PS4 in their videos.

You're giving them a shitload of excuses kevboard, its incredible.

spoken like someone who sees everything through the lens of console wars, which it seems is the main reason people take issue with what DF say. you have to have some Derangement where everything is a plastic box A vs plastic box B war, instead of just a simple tech analysis.
it's always looked at through fanboy glasses.

During the course of the years DF has shared a LOT of questionable opinions about the Switch 2 with a lot of "because Nintendo".

no they haven't.

Unlikely that tensor cores will be able to have DLSS on that portable.

they never said that. once again a purposeful misunderstanding or down right nonsense statement.

Unlikely that it'll have VRR or 120Hz.

it was unlikely, most people didn't expect that leak to be true. and it was a simple statement done during 1 section of 1 podcast where none of them tried to make any serious statement or prediction.

Would be less powerful than a Steam deck.

source? never heard them say that.

Don't believe the quality of the matrix demo.

well yes... what? the supposed people that saw it said it looked on par with the Series X version of the demo... that's just not possible, not even remotely.

For 3 weeks they declared it would be PS4 target while omitting the narrative that of SF6

what does that even mean? the grammar isn't garmmaring. SF6 uses essentially PS4 settings... so...

or that Cyberpunk 2077 phantom liberty is simply not running on PS4.

they didn't omit that and in fact always made clear that the Phantom Liberty parts are more demanding than the main game when the Switch 2 version was discussed. which you clearly blocked out with your fanboy glasses.

I/O streaming and more recent architecture, more powerful CPU? Eh who cares.

they constantly talk about these being the aspects that will make it punch above its weight... almost every time they talk about multiplat titles.

Why try to put some context into what we're seeing? FF7 intergrade with upgraded lighting? Eh, same as PS4 clearly.

there was no lighting upgrade, there was a lighting change. it's mostly cosmetic and has no real performance impact. image quality and performance are closest to PS4 and not any other version.

RDNA 2 Steam deck? OMG WOW! Portable Series S??

they never said that. they simply showed how Series S caliber titles run on the Deck, which uses the same exact architecture.

They were OVERLY optimistic for Steam Deck at launch.

were they? where? it always was the 720p PS4 every time they talked about it.
but similarly to how they talk about the Switch 2, they then also qualify that statement by highlighting the aspects of the Deck that are more modern than the PS4.

They are OVERLY conservative about Switch 2

how? they literally had the near exact specs of the system for 2+ years, and everything that was expected basically came true.

Both takes hurts their credibility.

once again only through the console war "how dare they not praise my favourite box more!" glasses

Metaldave exposes them as having very different posture. DF realized maybe after 3 years that their claim of portable Series S is crumbling and that's also not a good look on them to be honest,

again, they never said that. deliberate misinterpretation of fanboys.

they seem completely incompetent in what they are supposed to do just like I've also showcased how utterly incompetent they are with Switch 2 target with their stupid as fuck emulated 2050 GA10x which is not even at the time of cutdown T234 the same SM as GA10b is with the double FMA tensor cores like GA100 AI chipsets and they made an entire freaking video trying to gauge DLSS performance on a downclocked 2050 when its not even remotely close to AGX Orin. How can these peoples not even bother to open a whitepaper when they're spewing this shit to a large audience?

half of that is incomprehensible... english please? commas and periods would help you know?

as for the parts I understood,
they never said the 2050 is a perfect comparison. however the DLSS tests have yet to be disproven.
which games have we seen use DLSS? games that run below 1080p and upsample to 1080p or 1440p, and the latter only at 30fps.

their entire argument was that DLSS from 1080p to 4k takes up too much frametime to target 60fps on a chip of that class... that's literally all they said. have they been disproven? if so, show me.
they did these tests because people back then saw DLSS as a free performance boost, when in reality you can't just throw DLSS on everything and expect it to be 4k like or make everything 60fps. that was the entire purpose of these demonstrations and nothing else... and they are correct.

we have seen 30fps games targeting 1440p with DLSS, and 60fps games targeting 1080p with DLSS.
both still in line with that they said.

This is the reference of tech news on youtube for consoles?

better than super "must be advanced SSR or RT" dave 64, that's for sure.
the retard that uniroincally says shit like "DLSS to 1080p and then upscaled with TAA to 4k"

They've become a joke. Both Steam deck launch take and now Switch 2 take is not shining a good light on them*

*if you are a fanboy with console war glasses on

You keep trying to say we're too stupid to understand the context, are you for real?

yup, you are. or you misinterpret things on purpose, as that's a typical fanboy move.

I think pretty much everyone here can understand a more modern architecture with better CPU and more VRAM and decompression engine making it run games that devs gave up to make it run on base PS4 makes it beyond PS4 baseline.

and DF is stating this every time when they do compare it to the PS4. it's always said.
it's always something along the lines of "roughly PS4 ballpark but with advantages that make it punch above its weight"

Yet DF refuses to even acknowledge it.

they do... unless your fanboy glasses block it out.

Its not Series S either like I said, but their narrative that its baseline PS4 is completely stupid. If you believe this you're just actually too stupid to understand context.

it has roughly the same raw GPU power of a PS4. it's not exactly identical, but closer to PS4 than PS4 Pro.
their narrative is that it's in this ballpark + better CPU + modern features that allow modern ports + faster storage.

once again, you can listen to them constantly say this, but you deliberately chose to ignore that because they aren't overselling it, which the fanboys want them to do.

it's very clear that every time they aren't just mindlessly praising everything about the thing you want them to praise, their words get misconstrued into them "downplaying" or "being unfair".
and on top of that retards like SMD64 are grifting idiotic fanboys that can't tell the difference between a canned animation of Ryu's arm and the muscle deformation layer found on PS5 and Series X.

 

[Zathalus]

You're 2 upper tiers of graphic cards above 3060. Like really?

All your subsequent numbers are wrong TF for 3060.

NVIDIA GeForce RTX 3060 12 GB Specs

NVIDIA GA106, 1777 MHz, 3584 Cores, 112 TMUs, 48 ROPs, 12288 MB GDDR6, 1875 MHz, 192 bit

www.techpowerup.com

Do you not know how conservative the official spec sheet is for Nvidia GPUs ever since Turing? The official number is 12.74 TF and that is derived from the clock speed of 1777mhz. No 3060 actually operates at that speed.

EVGA GeForce RTX 3060 XC Review

EVGA engineered a compact dual-slot design with the RTX 3060 XC that will fit all cases. Unlike all other RTX 3060 cards we've tested today, a metal backplate is included. EVGA's card ticks at a rated boost of 1852 MHz, and the cooler features the idle-fan-stop capability.

www.techpowerup.com

Median of 1912Mhz

MSI GeForce RTX 3060 Gaming X Trio Review

The MSI GeForce RTX 3060 Gaming X Trio uses a huge triple-slot, triple-fan cooler that's much better than any other RTX 3060 we've tested today. Unsurprisingly, temperatures and noise levels in our RTX 3060 Gaming X Trio review are outstanding, with it runing incredibly cool while being...

www.techpowerup.com

Median of 1912Mhz

Palit GeForce RTX 3060 Dual OC Review

Palit's GeForce RTX 3060 Dual OC is a compact dual-slot design that comes with a factory overclock to 1867 MHz boost. It is the only card we tested today that has its default power limit raised beyond the NVIDIA default setting of 170 W, which should help with performance.

www.techpowerup.com

Median of 1935Mhz

Zotac GeForce RTX 3060 AMP White Edition Review

Zotac's RTX 3060 AMP White Edition offers a refreshing change in color theme that will appeal to many who want to show off their hardware in a case with a window. The card comes overclocked out of the box, with a very good dual-slot cooler that achieves excellent temperatures and decent noise...

www.techpowerup.com

Median of 1927Mhz.

I used 1927Mhz, but 1911Mhz just drops it down to 13.7 TF.

I mentioned the Ampere clock speed in my post, apologies I didn't go into further detail regarding how widely the official spec and real world clocks differ.

And why would devs making impossible ports on Switch ignore its best assets again?

They won't? But that's not what our discussion is about is it? We are talking about the raster performance here. Throw in RT and the Switch 2 should really outperform the Steam Deck.

I didn't pick RTX setting, its software RT

Since when RDNA 2 not performing well with software RT? Where does nvidia RT hardware even provide better performances with software RT? You understand software RT right?

I know what Software RT is. The Snowdrop engine uses fully accelerated hardware ray tracing.



Both RDNA 2 and Ampere are hardware accelerated.

Holy shit where you keep pulling these numbers 16.2 TF

NVIDIA GeForce RTX 3060 Ti Specs

NVIDIA GA104, 1665 MHz, 4864 Cores, 152 TMUs, 80 ROPs, 8192 MB GDDR6, 1750 MHz, 256 bit

www.techpowerup.com

I've already explained. But once again (using the founders edition no less):

NVIDIA GeForce RTX 3060 Ti Founders Edition Review

NVIDIA's GeForce RTX 3060 Ti comes at incredible pricing of $399, yet beats the GeForce RTX 2080 Super in rasterization, and even the RX 6800 in raytracing. In our RTX 3060 Ti review, we found that NVIDIA picked a fantastic cooler that's extremely quiet, yet fits everywhere because of its...

www.techpowerup.com

Median clock speed is 1875Mhz, getting us to 18.24 TF.

3060 Ti 16.2 TFlops just a tad under the 6800 16.2 TFlops

It's not, you can check the advertised vs actual clock speed of the 6800 if you wish:

AMD Radeon RX 6800 Review

The Radeon RX 6800 is the world's most power-efficient graphics card, better than anything NVIDIA Ampere has to offer. The smaller brother to the RX 6800 XT comes with a smaller heatsink, at more affordable pricing, but retains the lavish 16 GB VRAM memory buffer.

www.techpowerup.com

AMD is not nearly as conservative with boost clocks advertising like Nvidia is.

This is the peoples claiming that we have to halve Ampere TFlops?

Seems RDNA 3 we have to halve TFlops

I never claimed you have to halve Ampere, as I mentioned it is actual dual issue 32 compute. RDNA 3 uses VOPD to get those numbers, a totally different thing that no game actually uses.

AMD Radeon RX 6800 Specs

AMD Navi 21, 2105 MHz, 3840 Cores, 240 TMUs, 96 ROPs, 16384 MB GDDR6, 2000 MHz, 256 bit

www.techpowerup.com

NVIDIA GeForce RTX 3060 Ti Specs

NVIDIA GA104, 1665 MHz, 4864 Cores, 152 TMUs, 80 ROPs, 8192 MB GDDR6, 1750 MHz, 256 bit

www.techpowerup.com

NVIDIA GeForce RTX 3070 Ti Specs

NVIDIA GA104, 1770 MHz, 6144 Cores, 192 TMUs, 96 ROPs, 8192 MB GDDR6X, 1188 MHz, 256 bit

www.techpowerup.com

Yes, actual clock speeds and listed clock speeds can differ quite a bit.

Why? What's the excuse now?

Maybe it was build for PC in mind? Not a slop console port build by AMD → PC?

No excuse? As mentioned different games can widely different showings on different architectures. Some favor AMD, others favor Nvidia. Hence, averages are good to use. The more, the better.

With the wrong TF all over the place again

NVIDIA GeForce RTX 3080 Specs

NVIDIA GA102, 1710 MHz, 8704 Cores, 272 TMUs, 96 ROPs, 10240 MB GDDR6X, 1188 MHz, 320 bit

www.techpowerup.com

Nope.

NVIDIA GeForce RTX 3080 Founders Edition Review - Must-Have for 4K Gamers

NVIDIA's new GeForce RTX 3080 "Ampere" Founders Edition is a truly impressive graphics card. It not only looks fantastic, performance is also better than even the RTX 2080 Ti. In our RTX 3080 Founders Edition review, we're also taking a close look at the new cooler, which runs quietly without...

www.techpowerup.com

1950Mhz median.

So a dev that would make a game specifically for Switch 2 would use better the GPU with NVapi and nvidia SDKs than PC

Why are peoples making the narrative that you HAVE to turn down TFlops on Ampere architecture when I clearly showed that its game dependent?

Does it happens sometimes? Yes. On PC.

Would it make sense to happen on a closed platform with all the Nvidia tools to work on Ampere? No.

Because going from INT32+FP32 to FP32+FP32 or/and INT32+FP32 obviously has benefits, but it's certainly not going to result in a full doubling of performance.

Thanks to PS5? What is this? Low level APIs have proven themselves way before PS5.

Sure? I was just using the PS5 as the latest example. I could mention assembly or the PS2/PS3 era, but I didn't think it would be that relevant.

Really? Go build a Jaguar mobile CPU + 8GB + Tahiti HD 7800 and come back to us with running Horizon Forbidden west.

Horizon Forbidden West is not third party.

Nintendo will never invest huge budgets for that, so no.

You don't need that much of a budget to push the hardware to the limits. Luigi's Mansion on the Switch comes to mind.

You'll never catch DF ever saying anything positive of the kind for sure

They have said quite a bit that is positive about the Switch 2, but everybody just seems to ignore that.

I'd hope so? That's the point of picking Nvidia. You're also playing old demo.



How would you even detect internal res at a show. The end result is upscaled by AI. DF could not even detect DLSS for 3 weeks.



*Specific DLSS model made expressively for Switch 2. Challenging even transformer models.



Switch 2 is reportedly in medium/high settings for Cyberpunk 2077, unlike all the "low" steam deck showcases when they try to make the game run above 20 fps in dogtown.


At low settings and XeSS has never showcased like Nvidia's new DLSS that its a good thing to have at low resolutions.

Switch 2 DLSS wipes the floor with DLSS CNN model. Challenging even transformer model at low res.

I have no idea what the pixel counts were, just that it is obviously low in order to get those high/medium settings. Just look at offscreen footage from the latest DF video here (ideally fullscreen on a decently large screen at 4k to get a good idea):


No real point in discussing this one further, I don't really have any video evidence apart from what is already on YouTube. The pixel counts will be out in a few weeks anyway.

That all aside, where are you getting the double FMA tensor cores for Orin from? The Nvidia Jetson Whitepaper?

 

[HRK69]

Keep it going lads

 

[Azelover]

I actually enjoy SuperMetalDave quite a bit..Digital Foundry not so much...

He does good content for his audience and has been doing it for many many years. I think he deserves to have a little more visibility.

 

[Zathalus]

Keep it going lads

 

[GigaBowser]

I actually enjoy SuperMetalDave quite a bit..Digital Foundry not so much...

He does good content for his audience and has been doing it for many many years. I think he deserves to have a little more visibility.

metaldave nort even bragging bouts Switch 2 HE IS JUST LAUGHING AT DF SHILL FOR SAYING STEAM DECK IS SERIES S EXPERINCE IN 2022 OOHOOHOOHOOHOO

 

[Buggy Loop]

They won't? But that's not what our discussion is about is it? We are talking about the raster performance here. Throw in RT and the Switch 2 should really outperform the Steam Deck.

Or even raster. Again, I'll bet on it if you want.

I know what Software RT is. The Snowdrop engine uses fully accelerated hardware ray tracing.

Because going from INT32+FP32 to FP32+FP32 or/and INT32+FP32 obviously has benefits, but it's certainly not going to result in a full doubling of performance.

Of course. But "impossible ports" will make use of it the most. Games with mesh shaders come to mind.

Horizon Forbidden West is not third party.

Almost nothing on PC is dedicated to it. Take that PC build from 2012 and run Detroit become human on it then. It won't be pretty. It's not negligible what a closed platform does for efficiency compared to PC platforms.

So Ampere will shine even more so on its own dedicated metal API on a closed platform.

You don't need that much of a budget to push the hardware to the limits. Luigi's Mansion on the Switch comes to mind.

The art style fits a certain mid graphic budget but they'll never push the console like Sony pushed PS4. There's never gonna be a TLoU part 2 or Uncharted 4 effort by Nintendo.

That all aside, where are you getting the double FMA tensor cores for Orin from? The Nvidia Jetson Whitepaper?

Post in thread 'Nintendo Switch 2 will have the best upscaling technology of any consoles'

Apr 4, 2025

I doubt Switch 2 have even quarter of PS5Pro TOPS

And I really hope that PSSR does not use all of PS5 Pro's TOPs, because otherwise the tech is a joke.

(and no, I don't think it uses all TOPs, this has been discussed many times)

DLSS 4 is running on Turing cards that are in under 100 TOPs range and of course its not using all tensor cores all at once. Never in a render graph you'll see 100% utilization for upcaling.

How many tensor cores do you really think this thing...

• Buggy Loop

• 
• 

 

[TheKratos]

Metal GEAR?!

 

[Zannegan]

CONFIRMED

Metaldave? The same guy whose sources "confirmed" the Nintendo NX was a trafitional home console and NOT a hybrid? Watching his live reaction to the reveal was glorious beck in the day.

 

[GigaBowser]

 

[rodrigolfp]

Dude, even steam deck players that played cyberpunk 2077 are all saying switch 2 footage looks leagues better. Stop trying to make Steam deck cyberpunk 2077 dogtown a thing. Its not.


Never gonna be the entertainment of seeing how retarded one can be to halve Ampere TFlops

Did they played modded with XesS? Because with FSR it looks like shit, yes, plus not limited at 30fps shit show.

 

[GigaBowser]

Did they played modded with XesS? Because with FSR it looks like shit, yes, plus not limited at 30fps shit show.

please no my friends steamdecks is crud in compared thirs is why metal dave laugh

 

[rodrigolfp]

please no my friends steamdecks is crud in compared thirs is why metal dave laugh

And go above 30fps while even costing less.

 

[Buggy Loop]

indeed, which is why they made videos comparing the Deck with the PS4, even basically calling it a 720p PS4 in their videos

spoken like someone who sees everything through the lens of console wars, which it seems is the main reason people take issue with what DF say. you have to have some Derangement where everything is a plastic box A vs plastic box B war, instead of just a simple tech analysis.
it's always looked at through fanboy glasses.

DF is not even doing tech analysis, FFS, they let peoples from discord with god knows what footage with filesizes in the 100MBs range rather than Gigabytes of typical marketing material, to do pixel counts for them. Not DF, a fan! And they report that as news.

Or their internal evaluation where they pick the sole time a promo video goes from fullscreen to a render of the handheld not even taking all the screen and they pixel count and report that as the minimum resolution. What kind of journalism is this?

Who in their freaking mind would pixel count this and report on it?

they never said that. once again a purposeful misunderstanding or down right nonsense statement.

See at the bottom of reply

it was unlikely, most people didn't expect that leak to be true. and it was a simple statement done during 1 section of 1 podcast where none of them tried to make any serious statement or prediction.

Oh ok, they aren't serious. Good to clear that up.

source? never heard them say that.

they constantly talk about these being the aspects that will make it punch above its weight... almost every time they talk about multiplat titles.

They've made no efforts to say it. There's pretty much only John on social media that kind of says it'll punch above its weight. None of the Rich / Oliver / Alex really do.

there was no lighting upgrade, there was a lighting change. it's mostly cosmetic and has no real performance impact. image quality and performance are closest to PS4 and not any other version.

So they decided to not bring to PS4 because... ?

Lighting on PS4 version does not even impact dynamically on character

In fact a ton of lights were simply without impact at all with any of the environment.

DF : "While new light sources now cast coloured light against passing NPCs. PS5 finds itself a real demo moment here"

"there was no lighting upgrade," - Kevboard

DF : "Square enix uses PS5's three times boosting gpu power over PS4 Pro to push thicker layers of fog across multiple areas in the game"

DF : "Where PS5 shines especially is in the added light points though which illuminate areas previously shrouded in shadow. ... In fact even shadow quality itself it boosted across the game to deliver clearer outlines.

Switch 2 has same fog. Same lighting. We see the texture overhauls. So DF was just out of context here again right? With PS5 GPU being used to push these effects just out of context? I mean we have to excuse them. Anytime they open their mouth it seems we're not in the same world when it comes to Switch 2. PS5 is WOW, GPU pushing hard. Switch 2? EEHHHH? Just like PS4 basically. PS4 could do this obviously. No cost in performance. Let's bring in James while we're at it kevboard, its because they didn't think it would be worth it to upgrade the biggest userbase during that timeframe right? No money incentive?

they never said that. they simply showed how Series S caliber titles run on the Deck, which uses the same exact architecture

DF : "But when Steam deck is firing on all cylinders it can absolutely achieve the portable series s dream"

how? they literally had the near exact specs of the system for 2+ years, and everything that was expected basically came true.

With docked barely above Steam deck level as per DF in the video I included above? Under steam deck portable?

Dude

once again only through the console war "how dare they not praise my favourite box more!" glasses

See above. Oliver's experience with Steam deck is still making him gush over it. Somehow Steam deck is just magic, who needs closed platform, metal API? Flop for Flop it beats a Switch 2 handheld with windows/Linux+proton with shit directX 12 by DF's very own Oliver.

But sure, its all because of a box that I'm questionning them
I've nailed almost all switch 2 predictions so far and never went overboard. Even when the likes of MLID was starting narratives that it would be faster than series S i've always brought back that argument down.

again, they never said that. deliberate misinterpretation of fanboys.

DF : "But when Steam deck is firing on all cylinders it can absolutely achieve the portable series s dream"

and DF is stating this every time when they do compare it to the PS4. it's always said.
it's always something along the lines of "roughly PS4 ballpark but with advantages that make it punch above its weight"

Pretty much only John says that

As they were pixel counting a shrunk video to a portable render for 540p, Mr Oliver ...

DF : "But this is a really helpful reality check because I've seen a lot of discussion online, a lot of people debating this, especially after we published last week's DF direct which was about this topic to some substantial degree and you know the evidence that we have, that's high quality evidence like our direct comparisons with our pixel counts, its not gonna be this mega powerful thing, especially out of the gate. It's not going to be, you know, even a portable PS4 in handheld form.

Drastic departure from "it can absolutely achieve the portable series s dream" for a steam deck.

That's what peoples are calling out.

it has roughly the same raw GPU power of a PS4. it's not exactly identical, but closer to PS4 than PS4 Pro.their narrative is that it's in this ballpark + better CPU + modern features that allow modern ports + faster storage.

See above Oliver's prediction just a month ago...

once again, you can listen to them constantly say this, but you deliberately chose to ignore that because they aren't overselling it, which the fanboys want them to do.

it's very clear that every time they aren't just mindlessly praising everything about the thing you want them to praise, their words get misconstrued into them "downplaying" or "being unfair".
and on top of that retards like SMD64 are grifting idiotic fanboys that can't tell the difference between a canned animation of Ryu's arm and the muscle deformation layer found on PS5 and Series X.

Or they spend again a long portion getting the 2050 out and downclock it even further with visually a game that looks nothing like what we see from Switch 2 showfloor, as it looks like dogshit even in base game, not dogtown, base game.

And they again make the argument, because they didn't think DLSS was in any games thus far, that anyway its better to drop DLSS even at low res, better to do a TAA pass.

Even as they are testing for 720p, "I actually think the cost of DLSS is such that probably a TAA upscale with a better internal resolution is the way forward there". Not just for 4K claim like you did previously. These guys did not believe it made any sense for this handheld to use DLSS because of their fucking 2050 they keep bringing up time and time again.

How did that age? We have DLSS cyberpunk 2077 and the quality of it shames CNN model and XeSS.

 

[Deerock71]

I didn't want to post another thread, but now DF is comparing Switch 2 to PS5 in performance mode. I do wish they would be consistent.

 

[ResilientBanana]

Man, OP really thought he had a mic drop moment backing a complete and utter shill with no technical background.

 

[Buggy Loop]

Man, OP really thought he had a mic drop moment backing a complete and utter shill with no technical background.

Why peoples keep attacking Dave?

He doesn't need to have sources. He doesn't need pixel counters to look at what DF is saying in one video compared to another one. Anyone can pull up a DF video and do the same.

 

[thicc_girls_are_teh_best]

comparing 2 home consoles that will share many games between them makes no sense?

That's not the basis DF used when comparing them, though. And it's not like they will really share too much of the same library; most pubs have moved on from PS4, whereas Switch 2 is brand new. Yes some of the biggest multiplats on PS4 that skipped Switch might get Switch 2 releases, but we're probably talking about like 10% of the PS4's library, at best.

they also never compared the hardware, they compare roughly what the quality of the graphical presentation of games will be like. and they will be PS4 like.

Except they can't know what that will look like in actual effect, because Switch 2 hasn't even officially launched yet. Which also means, it hasn't been on the market long enough for devs to optimize development pipelines around the system, the way they've managed to for the PS4.

Also different studios have different aims in the games they make. 3P games, most of the new ones anyway, coming out this gen will be targeting PS5 as the main platform due to install base size. It's a good chance some of those studios won't even main the Switch 2 versions of those games once they port those over. That will have an impact on versions of those games to other platforms, like Switch 2.

Street Fighter 6 is essentially using PS4 settings for example

Yet it looks better than the Series S version in most instances.

it once again just clearly seems to be the case that people either purposefully try to misinterpret what DF says or are just actually too stupid to understand context

Maybe if DF themselves knew what they were saying more often, people wouldn't misinterpret them?

 

[nkill20]

 

[kevboard]

That's not the basis DF used when comparing them, though. And it's not like they will really share too much of the same library; most pubs have moved on from PS4, whereas Switch 2 is brand new. Yes some of the biggest multiplats on PS4 that skipped Switch might get Switch 2 releases, but we're probably talking about like 10% of the PS4's library, at best.

Harry Potter, SF6, Cyberpunk, Fortnite FF7... and that's only the launch window.

Except they can't know what that will look like in actual effect, because Switch 2 hasn't even officially launched yet. Which also means, it hasn't been on the market long enough for devs to optimize development pipelines around the system, the way they've managed to for the PS4.

they literally say this as well. they compared the launch window titles with overlap and said that it will only get better from there. so once again selective perception. positive statements get ignored or forgotten, and "negative" ones get highlighted. I put negative in quotes as that's only the case if you are a fanboy, otherwise it's just neutral observations

Also different studios have different aims in the games they make. 3P games, most of the new ones anyway, coming out this gen will be targeting PS5 as the main platform due to install base size. It's a good chance some of those studios won't even main the Switch 2 versions of those games once they port those over. That will have an impact on versions of those games to other platforms, like Switch 2.

ok... sure...

Yet it looks better than the Series S version in most instances.

because the Series S version looks worse than the PS4 version in some ways lol.

it has better textures due to larger ram pool, the rest is PS4 settings. no muscle deformation layer, no moving pebbles and debris, no SSR.

the series S version is also largely equivalent of the PS4 version. the big and almost only advantage it has is that story mode has a performance mode, which is missing on PS4/pro

Maybe if DF themselves knew what they were saying more often, people wouldn't misinterpret them?

they know what they're saying. misinterpretations only happen because of people being selective of what they acknowledge they said, and what they ignore of what they said.

 

[GigaBowser]

And go above 30fps while even costing less.

whaaatt?? steemdeck almost same price 720p low setings you must turn off trees to run cyberpunks ahahaha

 

[KeplerL2]

You guys are spending way too much time discussing something that doesn't really matter. The difference in power between PS4/Steam Deck/Switch 2 is about the same as the difference between PS5/Xbox Series X, aka not enough to make a noticeable difference most of the time.