Fake
Member
Is wired second article in the first page?Okay boys send me all the new info you want so that I can update my post on the first page
Is top priority.
Is wired second article in the first page?Okay boys send me all the new info you want so that I can update my post on the first page
Lemme put some stuff togetherIs wired second article in the first page?
Is top priority.
Lemme put some stuff together
Cripple their ecosystem? If anything it would strengthen it. The more people playing PlayStation games the better and the stronger the community for each game is. Why stop at 100,000,000 consoles? Let's not forget that they will most likely double down on Playstation Now next-gen considering their competition is more than Microsoft in that space. For them to simply disregard that would be pretty arrogant.Who ever said indirect competition never hurt Sony? Of course the PC is competing with consoles because ummm it plays games and very well I might add on top of being cheaper. Let's not be so naive to the facts, many on this very forum would love for all of Sony's titles to come to pc. That would cripple their ecosystem and I for one love that ecosystem and it ceasing to exist would be detrimental to the gaming industry. Business 101 people, it's a very simple concept.
Hopefully this won't impact console prices:
Samsung Memory Production Impacted by Equipment Issue, Predicts Millions of Dollars in Damage
Samsung's DRAM production hits a snagwww.tomshardware.com
IIRC, platform holders do futures contracts to stabilize/lock down component prices.
I remember back in late 2017/early 2018 GDDR5 prices had increased, but the XB1X price had remained stable.
For all the coding to the metal discussion. I am developer but not a game dev.(Though I have wrote some raytracing and wrote my own 3d engine 19 years ago that ran quake 3 levels). However one of the most innovative, smartest, a true game dev expert amongst the top .1 percent and gifted 3d engine dev ever, John Carmack said this about consoles vs PC's :
...and that was written in 2014. Not saying there's no truth to it, there most certainly is. But what relevance does Carmack have to gaming nowadays really? Promoting Oculus?
Coding to a fixed spec has advantages, especially when developers have the time to get used to the hardware. Hell, see some C64 demos done today - they're absolutely mindblowing. Just that you using Carmack as a reference, in this day and age, is quite yawn-inducing.
(And really, Ultima Underworld had a better engine than Doom and it was released over a year before it. I've said it before and I'll say it again, Ultima Underworld is very much underrated.)
More possibly coming true of that 14TF rumour that was deleted recently.
I'd put my opinion on the line saying that i guarantee his current knowledge wipes the floor with any poster on this message board, and most current game devs, about 3d engines and game development software and hardware....besides Tim Sweeny, Unreal engine Mastermind has said similar things about consoles and PCs. 2014 used amd-64 and that cpu Arch is what we still use today as well as gpu hardware is completely similar to 2014, see Xbox 1x etc.. and ps4 and latest gpus, the only big advancement is raytracing hardware acclereation..not raytracing itself...see my post above, I wrote raytracing software in 2000 and the concepts are similar today, mostly revolves around intersections, the math and calculations have been around for ages. The biggest change , catching up for PC's to consoles in some regard, has been Mantle, DirectX 12 and Vulkan in regards to draw calls...but that's not enough to close the gap too much that was stated by Carmack in 2014
I beg to differ on the authority part...at least if it's "good" authorityCouple of things.
1. Appeal to authority is not a good argument.
2. I agree that probably 2x PC hardware is a good approximation. But probably it's smaller.
3. Ray tracing nowdays is all about de-noising.
Almost none of us has the resources , time or expertise (I'm not saying capability) to research and thoroughly make evident the truth on this subject and present it here. So an appeal to expertise is wise, cost saving and most probably accurate. If you want to benchmark exact, scientific like for like which would be the only better way than an trusted expert who's seen it themselves firsthand by their own experience, then be my guest, I expect your research , benchmarking analysis report when? That be the only other evidence that would trump an experts expertise would be an objective research paper/study, you'd even be able to publish it in a computer science journal I'm sure
Sure but we have Carmarks roughly 2x, non exact, but striking performance number, which if off 1-5 percent or whatever margin of error that's realistic...it's pretty remarkable. So yeah not exact, but roughly close in terms of relative, overall performance metric results, maybe closer to 110 percent more or 90 percent, rough, but still amazingAny comment on console specifics is under NDA. Therefore there's no way to have exact numbers.
I'd put my opinion on the line saying that i guarantee his current knowledge wipes the floor with any poster on this message board, and most current game devs, about 3d engines and game development software and hardware....besides Tim Sweeny, Unreal engine Mastermind has said similar things about consoles and PCs. 2014 used amd-64 and that cpu Arch is what we still use today as well as gpu hardware is completely similar to 2014, see Xbox 1x etc.. and ps4 and latest gpus, the only big advancement is raytracing hardware acclereation..not raytracing itself...see my post above, I wrote raytracing software in 2000 and the concepts are similar today, mostly revolves around intersections, the math and calculations have been around for ages. The biggest change , catching up for PC's to consoles in some regard, has been Mantle, DirectX 12 and Vulkan in regards to draw calls...but that's not enough to close the gap too much that was stated by Carmack in 2014
More possibly coming true of that 14TF rumour that was deleted recently.
Oh yes I know this one. I thought there were new rumors corroborating it.
Was after or before RDNA?
Oh yes I know this one. I thought there were new rumors corroborating it.
Was after or before RDNA?
Even if it's GCN 14TF then we are looking at Radeon 7 and RTX 2080 performance. With GPU power like this in a console we should expect some really stunning looking games, especially when developers will aim at 4K checkerboard and 30fps.It doesn't state GCN so it should just mean 14TF.
Don't you wish 30fps for next gen, it might happen!Even if it's GCN 14TF then we are looking at Radeon 7 and RTX 2080 performance. With GPU power like this in a console we should expect some really stunning looking games, especially when developers will aim at 4K checkerboard and 30fps.
Don't you wish 30fps for next gen, it might happen!
I think next gen will have enough improvements/power/TFs for Dynamic/CB 4K, ray tracing and 60fps. That's the minimum I'm expecting.
At the same I weaken at the thought of what ND could achieve locking the frame rate to 30fps.
I'm in and endless loop of graphics vs performance whoring, I wish we could have both.
I saw. Pretty sweet. @Mista help me out in that wired thread too. At least people can forget about 'PS5 software RT' once and for all.hey guys i started to update the second post. thankCrustyBritches for all his hard work
Is really that hard to notice how fake that pastebin leak is?
For all the coding to the metal discussion. I am developer but not a game dev.(Though I have wrote some raytracing and wrote my own 3d engine 19 years ago that ran quake 3 levels). However one of the most innovative, smartest, a true game dev expert amongst the top .1 percent and gifted 3d engine dev ever, John Carmack said this about consoles vs PC's :
Added something similar to post #2. Thanks for the reminderDigital representations of the cartridges.
![]()
More in the link.
![]()
Sony PlayStation 5 met uitbreidbaar SSD geheugen cartridge | LetsGoDigital
Zal de Sony PlayStation 5 met cartridges werken om het SSD geheugen uit te kunnen breiden? Een onlangs gepubliceerd design patent suggereert van wel.nl.letsgodigital.org
hey guys i started to update the second post. thankCrustyBritches for all his hard work
What did I doI saw. Pretty sweet. @Mista help me out in that wired thread too. At least people can forget about 'PS5 software RT' once and for all.
I think he's mainly referring to CPU/RAM and maybe GPU in an overall picture sort of deal. I just tested Ryzen 2c/4t @ 3GHz(Ryzen master won't allow me to drop any lower, and I don't have old shitty CPUs anymore) plus RX 480(911MHz core, 1750Mhz mem) at PS4 Pro settings and it does just fine in comparison. On Pro you'll see dynamic res as low as 85% of 1440p, whereas on PC you can't replicate that except by testing at 1440p, then if the scene is too heavy dropping resolution scale. This is at 1440p...
*Please ignore the poor PS4 Pro image quality. I downloaded the DF vid, but it only provides .mkv for 1440p/4K, and I don't have anything that could convert it, so I had to go with 1080p. The framerate counter is the important part anyway.
DOOM makes use of Async compute, shader intrinsics, and frame flip optimizations according to AMD's detailing of it's Vulkan implementation. Having a better CPU on PC is usually a given. Although, I don't think the context of these optimization arguments are based on getting away with a lesser CPU while matching or under-performing the GPU.
Polaris in PS4P is no way full fat RX 480, far from that because modified RX 580 in xbox x GPU can sometimes double it's performance.
BTW- Doom wasnt so demandig on PC, however I wonder if RX 480 can match PS4P results in RDR2. This game is extremely demanding.
They have RX 580 results in 4K, RX 580 22fps min and 25 average, while xbox x run this game at solid 30fps (so without v-sync probably more than 35 fps on average). Of course these are not exact xbox x settings, but should be not that far at this point, while RX 580 on PC still need 30% performance more in order to match xbox x results. According to digital foundry benchmark chart currently you need vega 56 in order to play at 4K with solid 30fps.According to the Digital Foundry video, they were able to replicate (well almost) Xbox 1X settings and RX580 had almost 60 fps locked @1080p. I don't remember if they tried this card @4k though.
PS4 Pro is indeed a full fat RX 480 with 36CUs, 911MHz core clock, and 1700MHz mem.Polaris in PS4P is no way full fat RX 480, far from that because modified RX 580 in xbox x GPU can sometimes double it's performance.
BTW- Doom wasnt so demanding on PC, however I wonder if RX 480 can match PS4P results in RDR2. This game is extremely demanding.
It should be noted that it shares some bandwidth with the CPU and I went with 1750MHz, but there's only so far down you can take the card with memory clock before it begins to behave oddly.
DOOM is relevant because the subject is John Carmack and console optimization. It is also probably the closest thing the PC has to "coded to the metal"(async compute, shader intrinsics, frame flip optimization). Lastly, it was PS4 base's best-case scenario against the R7 265, likely because it had 8 ACEs compared to the 7850(R7 265)/7870/Xbox One which have 2. It took R7 265 with oc to 1100MHz to compete with PS4 base, and it still lacked the framerate consistency.
Xbox One X has cache setup more in line with Vega than Polaris(increased render cache, likely coherent pixel/texture memory access) , and higher memory bandwidth. This is why a 6TF Polaris card struggles in certain titles(DOOM, RDR 2) in comparison, and why X1X exhibits such an advantage over PS4 Pro(which chose Vega's RPM as forward feature).
For Xbox One X DOOM DF states:
-only Xbox One X that's capable of hitting native 3840x2160, and even then only in select circumstances.
-most common resolution we found in Doom is 3072x2160
-We've seen Doom on Xbox One X drop to 2880x1620 (a 75 per cent scaling on both axes)
-it can actually drop even lower than that
-At their lowest points, Xbox One X pushes 25 per cent more pixels than the Pro, but the gap in average gameplay rises to a circa 2x increase
For RDR 2, I don't have it yet and won't until it hits ~$25, but I'd guess PS4 Pro and RX 480(911MHz core, 1750MHz mem) will again be a match. RX 480/580 has a harder time against X1X to overcome the memory pipeline alterations and increase bandwidth of the X. That's why even with fat OC and the lowest matching settings from DF, it still will fall a bit short(maybe 5-10%).
tldr: This is hardware customization as opposed to software optimization. The memory configuration is where the honey is at, imo.
With so drastically downclocked GPU speed and limited memory bandwidth this thing in PS4P should be no longer compared to RX 480, although both GPU's has 36 CU's. As you can see doom runs clearly better on your RX 480 compared to PS4P version even with downclocked GPU speed.PS4 Pro is indeed a full fat RX 480 with 36CUs, 911MHz core clock, and 1700MHz mem.
Go to the first page.What did I do![]()
Ah yes. That was nothing irmãoGo to the first page.