Next-Gen PS5 & XSX |OT| Console tEch threaD

Status
Not open for further replies.
I haven't seen Microsoft give us that number yet so I am questioning it.

There's really no need for the number games, let's just see what the games look and load like. MS's storage optimizations are geared more towards lowering the amount of data that needs to be delivered (via texture paging, direct AI image upscaling, etc.) rather than increasing overall IO, which is fitting since they have less raw data speed.
 
Last edited:
The XSX claims it can do the same thing, streaming from the SSD. Yes. The PS5 is faster. Yes, the streaming itself cannot be matched by a slower system. But consider what it means in practice.
Streaming is not a new thing. That's what's been done all the time from hard drives. But hard drives are slow. It has been stated clearly that the slow hard drive means filling up RAM with data for at least the next 30 seconds or so.

Now let's jump to the SSDs... So if the PS5 can load data in 0.3 seconds for example, the XSX would need 0.6 seconds, roughly speaking. The expected difference at this point is that the PS5 needs lower RAM allocation than the XSX, because the additional data that would need to be loaded in that 0.3 seconds, would simply be put in RAM beforehand instead. The XSX will have less efficient RAM usage, and at this point that is the only main known drawback in practice for the XSX regarding the whole SSD & I/O aspect. How significant it will be, once again, remains to be seen.
Yes that's right, the faster your streaming the less buffering allocation in ram needed, resulting in more ram you have for what's to be rendered in front of you.
 
At the start of the generation, the Xbox One OS was so barebones, of course they added features more quickly (and most of those catching up to PS4 OS). They only got it more feature-rich than PS4 in the last couple of years.
You might want to look back, because the PS4 OS was also barebones, Microsoft was just much faster in adding new features and way more transparent where they gave the community the possibility to vote for features they wanted.
 
What do you mean 'now'? They advertised that from the beginning.

Its little nuances. While one can argue its the same thing, but their wording is not. Sony has stuck with elimination of load times.


" our next-generation SSD will virtually eliminate load times "


" Developers will also be able to effectively eliminate loading times "


" They will showcase unparalleled load-times "

" effectively eliminates loading times "

" Game worlds are larger, more dynamic and load in a flash "

"But they didn't want the I/O system to be just about your games loading faster"

"The most noticeable of these is loading times, which will be greatly decreased thanks to the processing power of Xbox Series X "


I don't know about you, but there is a difference between 'eliminate load times' and these words, at least for me.
 
4,8GB/s + 50% Bpack = 7,2GB/s :)

This is playing silly games with the numbers. Actually thinking about the numbers..... On PS5 it seems Kraken takes 5.5GB/s and compresses by ~60% overall? = 8.8GB/s hence the 8-9GB/s figure. No way then does 2.4GB/s get 100% compressed to 4.8GB/s with Zlib (10% worse than Kraken?) so must already include what BCPack brings for textures? Even if this brought 50% you can't just add both together consecutively right!? These numbers games are a mess!

Someone please explain it so I understand.....
 
This is playing silly games with the numbers. Actually thinking about the numbers..... On PS5 it seems Kraken takes 5.5GB/s and compresses by ~60% overall? = 8.8GB/s hence the 8-9GB/s figure. No way then does 2.4GB/s get 100% compressed to 4.8GB/s with Zlib (10% worse than Kraken?) so must already include what BCPack brings for textures? Even if this brought 50% you can't just add both together consecutively right!? These numbers games are a mess!

Someone please explain it so I understand.....

2.4GB/s to 4.8GB/s represents 50% compression. Let's say you have 4.8GB of data and you compress that 50%, you'd have 2.4GBs of data.
 

$70B in debt


$5.8B in debt
"How do modern companies work in 2020"
 
This is playing silly games with the numbers. Actually thinking about the numbers..... On PS5 it seems Kraken takes 5.5GB/s and compresses by ~60% overall? = 8.8GB/s hence the 8-9GB/s figure. No way then does 2.4GB/s get 100% compressed to 4.8GB/s with Zlib (10% worse than Kraken?) so must already include what BCPack brings for textures? Even if this brought 50% you can't just add both together consecutively right!? These numbers games are a mess!

Someone please explain it so I understand.....
don't try to make sense of these numbers. those delusional fanboys just glued what ever numbers they could find.
 
2.4GB/s to 4.8GB/s represents 50% compression. Let's say you have 4.8GB of data and you compress that 50%, you'd have 2.4GBs of data.

Okay that makes sense but then why does PS5's 5.5GB's only become 8-9GB/s when it uses Kraken which is said to be 10% better? Numbers are being fudged somewhere!
 
Not quite, for the last time, probably ;-)

Raw: 2.4GB/s Vs 5.5GB/s
Typical: ??? Vs 8/9 GB/s
Max theoretical: 4.8GB/s | 6GB/s (for textures) Vs 22GB/s

You are all going to be shocked at how fast Sony's SSD is compared to MS's. The biggest difference is going to be in the typical numbers. And we don't know XSX typical numbers.
Your table is not aligned right. It should be:
2.4GB/s raw VS 5.5GB/s raw
4.8GB/s typical vs 8GB/s - 9GB/s typical
??? max vs 22GB/s max

It's actually not that clear what the 6GB/s figure is but if I had to guess, it's the BCPack typical. The final typical is 4.8 because Zlib lowers the average.

How are you getting an odd number using a base 2 multiplier? Has to be an even number.
It's not an odd number, it's in GiB which means there are 9 digits after that 1.
 
Last edited:
they copy - paste whatever sony is doing. what they are going to do if they don't have any identity...

They've been following Sony's blueprint to gain good graces since after the beginning of this gen. Good examples "Gears will be more open" like God of war/Uncharted, "We are giving devs freedom to create their games" Sony pushing tht narrative for years, "Fast Loading" Sony brought tht forth in initial reveal and Microsoft's reveal was a carbon copy just in video few weeks, later. "We are investing and expanding first party studios" Sony mantra since PS3.

There's more but you get the hint.
 
Your table is not aligned right. It should be:
2.4GB/s raw VS 5.5GB/s raw
4.8GB/s typical vs 8GB/s - 9GB/s typical
??? max vs 22GB/s max

It's actually not that clear what the 6GB/s figure is but if I had to guess, it's the BCPack typical. The final typical is 4.8 because Zlib lowers the average.
The 6 GB/s is the theoretical maximum throughput of the hardware decompression block. So it's a hard limit. The 4.8 is typical. Similar to how the 22GB/s is the maximum for the PS5 block.
 



tqgB8LA.png
 
The 6 GB/s is the theoretical maximum throughput of the hardware decompression block. So it's a hard limit. The 4.8 is typical. Similar to how the 22GB/s is the maximum for the PS5 block.
It can't be, it doesn't make sense to the compression ratio of Kraken, Zlib and BC if the typical is 4.8GB/s.
 
Okay that makes sense but then why does PS5's 5.5GB's only become 8-9GB/s when it uses Kraken which is said to be 10% better? Numbers are being fudged somewhere!

BCPack is rumored to be more performant at image compression by 10 or 15%. Kraken is generally around 35 -40% on images, those are the typical numbers that Sony went with. On ascii to binary compression, Kraken can be a stellar performer, zlib not so good there. Sony covers the spread in raw performance, they are champs in this area, any noise to the contrary is just that.
 
It can't be, it doesn't make sense to the compression ratio of Kraken, Zlib and BC if the typical is 4.8GB/s.

Seems pretty clear from the DF article:

Our second component is a high-speed hardware decompression block that can deliver over 6GB/s," reveals Andrew Goossen. This is a dedicated silicon block that offloads decompression work from the CPU and is matched to the SSD so that decompression is never a bottleneck

BCPack is rumored to be more performant at image compression by 10 or 15%. Kraken is generally around 35 -40% on images, those are the typical numbers that Sony went with. On ascii to binary compression, Kraken can be a stellar performer, zlib not so good there. Sony covers the spread in raw performance, they are champs in this area, any noise to the contrary is just that.

Interesting. Given this, and until we get more details, I'm going to assume the 4.8GB/s figure includes BCPack and Zlib.
 
Interesting. Given this, and until we get more details, I'm going to assume the 4.8GB/s figure includes BCPack and Zlib.

Looks like a representation of their overall average compression target, agreed. It would have been foolish to go with a more capable decompression block in combination with the SSD they chose. Not much good if you can't feed it.
 
Its little nuances. While one can argue its the same thing, but their wording is not. Sony has stuck with elimination of load times.


" our next-generation SSD will virtually eliminate load times "


" Developers will also be able to effectively eliminate loading times "


" They will showcase unparalleled load-times "

" effectively eliminates loading times "

" Game worlds are larger, more dynamic and load in a flash "

"But they didn't want the I/O system to be just about your games loading faster"

"The most noticeable of these is loading times, which will be greatly decreased thanks to the processing power of Xbox Series X "


I don't know about you, but there is a difference between 'eliminate load times' and these words, at least for me.

And then you see that in their official loading time tech demo they dont include a stopwatch because they dont want you to notice that it takes around 11s (pop ups include) to load a low resolution game.



U5Ds8VD.jpg
 
Last edited:
Its little nuances. While one can argue its the same thing, but their wording is not. Sony has stuck with elimination of load times.


" our next-generation SSD will virtually eliminate load times "


" Developers will also be able to effectively eliminate loading times "


" They will showcase unparalleled load-times "

" effectively eliminates loading times "

" Game worlds are larger, more dynamic and load in a flash "

"But they didn't want the I/O system to be just about your games loading faster"

"The most noticeable of these is loading times, which will be greatly decreased thanks to the processing power of Xbox Series X "


I don't know about you, but there is a difference between 'eliminate load times' and these words, at least for me.
It's mostly pr. May have to wait for launch to see how it all shakes out.
 
Seems pretty clear from the DF article:





Interesting. Given this, and until we get more details, I'm going to assume the 4.8GB/s figure includes BCPack and Zlib.
If your typical decompression rate is 4.8GB/s (which means the BCPack average decompression rate is much higher than 4.8Gb/s), your max is way over 6Gb/s. Your average max might be 6-7 GB/s, but not the BCPack decompression max. Even Zlib can hit x3 compression ratio here and there (7.2GB/s).
 
Last edited:
Your table is not aligned right. It should be:
2.4GB/s raw VS 5.5GB/s raw
4.8GB/s typical vs 8GB/s - 9GB/s typical
??? max vs 22GB/s max

It's actually not that clear what the 6GB/s figure is but if I had to guess, it's the BCPack typical. The final typical is 4.8 because Zlib lowers the average.


It's not an odd number, it's in GiB which means there are 9 digits after that 1.
Then why is it smaller then the binary 1 TB number. Should be larger not smaller.
 
You might want to look back, because the PS4 OS was also barebones, Microsoft was just much faster in adding new features and way more transparent where they gave the community the possibility to vote for features they wanted.

I'm pretty sure the PS4 had more features from release than the Xbox I'm not saying it has now but at the start it did .
 
And then you see that in their official loading time tech demo they dont include a stopwatch because they dont want you notice that it takes around 11s (pop ups include) to load a low resolution game.



U5Ds8VD.jpg



It's basically just loading the Xbox One X version of the game correct?

If the games visuals were upgraded to Series X levels wouldn't the loading times be longer?

I'm just assuming an increase in asset size when we move into Next Gen.
 
Its little nuances. While one can argue its the same thing, but their wording is not. Sony has stuck with elimination of load times.


" our next-generation SSD will virtually eliminate load times "


" Developers will also be able to effectively eliminate loading times "


" They will showcase unparalleled load-times "

" effectively eliminates loading times "

" Game worlds are larger, more dynamic and load in a flash "

"But they didn't want the I/O system to be just about your games loading faster"

"The most noticeable of these is loading times, which will be greatly decreased thanks to the processing power of Xbox Series X "


I don't know about you, but there is a difference between 'eliminate load times' and these words, at least for me.

U read way to much into this lol.

Both consoles can have zero load times perfectly fine if they choose for it. The only advantage that sony SSD has is faster load times in games that do require load times because of how they are designed.

People keep forgetting that seamingless worlds without load times have already been a thing since forever even on HDD's.

It's basically just loading the Xbox One X version of the game correct?

If the games visuals were upgraded to Series X levels wouldn't the loading times be longer?

I'm just assuming an increase in asset size when we move into Next Gen.

SSD is getting bottlenecked hard because the game doesn't take the SSD into account for shit.

What u see there is not the full speed of the SSD, its much like what PC's do. create 2-5x faster load times, because of faster access to data which is more directly. Which gives them that load speed. as its 11 vs 54 seconds. The SSD in the xbox series X is probably 50-100 times faster depending if you use compressed io throughput. Which will mean if the game was designed around that SSD u would see practical no load times. However the game probably can't handle that because it was never designed for it. The same as ac odyssey is capped at 40mbps load times on PC no matter the SSD.

They clearly just dropped a xbox one game on series x and demonstrated the gains without optimisations for it.

So not really, load times will get shorter actually on both boxes.
 
Last edited:
Then why is it smaller then the binary 1 TB number. Should be larger not smaller.
The smaller is the binary. Storage volume is advertised in GB but in the OS itself you see all the numbers in GiB. That's why every 1TB drive on the market is actually 931GiB. That's why XSX 1TB drive is actually 931GiB and PS5's 825GB is actually 768GiB.
 
Last edited:
Both are doing it, Sony can do it faster. I'm not sure how the faster streaming of the PS5 will eventually look like in non-exclusive games. I mean, the SSD can stream x2 more textures, models, and other forms of data but all that extra data needs to sit in memory and PS5 doesn't have more or faster memory than XSX. And the CPU and GPU also need to be able to handle all of that stuff in order to render it to screen, but PS5 doesn't have a more powerful CPU and GPU. So I guess we will have to wait and see how these things will affect next-gen, it's going to be really interesting.

Couple of things:
1. If streaming is faster it means less RAM usage. Not more.
2. Rendering a high quality assets is pretty much the same as rendering a low quality one. You can check the perf of any fan-made texture packs and the differences are negligible. Even if the texture size was increased 2x across the board (google the recent Witcher 3 re-tex).
 
If your typical decompression rate is 4.8GB/s (which means the BCPack average decompression rate is much higher than 4.8Gb/s), your max is way over 6Gb/s. Your average max might be 6-7 GB/s, but not the BCPack decompression max. Even Zlib can hit x3 compression ratio here and there (7.2GB/s).

Honestly don't know but I can't argue with what Andrew said and what he specifically talks about here. I'm pretty sure if 7.2GB/s was correct it would be the number we see in the specs instead of 4.8GB/s.
 
It's basically just loading the Xbox One X version of the game correct?

If the games visuals were upgraded to Series X levels wouldn't the loading times be longer?

I'm just assuming an increase in asset size when we move into Next Gen.

You should also assume a package that better utilizes the speeds offered by SSD.
 
4,8GB/s + 50% Bpack = 7,2GB/s :)
This is wrong for two reasons.

For one, 4.8GB/s is the compressed bandwidth, so it already accounts for compression. The RAW bandwidth is 2.5GB/s.
So he's applying the compression ratio to a compressed bandwidth.

Secondly, he's comparing it to the PS5's RAW bandwidth of 5.5GB/s which in reality is still faster than the Series X's SSD bandwidth with compression.
 
It's mostly pr. May have to wait for launch to see how it all shakes out.

I agree this is mostly PR. There is a thin line between PR and lying. PR has to adhere to laws / legal department.

My post was to show the difference between PR of two companies. MS circles around the word "eliminate load times" while Sony directly says "eliminate load times".
 
U read way to much into this lol.

Both consoles can have zero load times perfectly fine if they choose for it. The only advantage that sony SSD has is faster load times in games that do require load times because of how they are designed.

People keep forgetting that seamingless worlds without load times have already been a thing since forever even on HDD's.
They do have load times though. They're just hidden behind long walking sections, crawl spaces, elevators,etc. The PS5's SSD will be able to eliminate all of these kinds of things, and allow developers to do more with their games. Lots of developers have expressed how excited they are for this, and how this opens up much more possibilities in game design. That being said, as far as I know, we haven't seen anyone praising the XSX SSD yet. We'll have to wait and see how first party games differ this gen, and if the Xbox will be able to keep up in those aspects.
 
It's basically just loading the Xbox One X version of the game correct?

If the games visuals were upgraded to Series X levels wouldn't the loading times be longer?

I'm just assuming an increase in asset size when we move into Next Gen.
That is exactly my point. Some users say that the game was not optimized for that tech demo without providing any evidence about that claim, but they "forget" to mention that the game has low resolution assets and still it takes 11 sec to be fully loaded.

Now imagine loading a very complex and full 4K game :)
 
Last edited:
I agree this is mostly PR. There is a thin line between PR and lying. PR has to adhere to laws / legal department.

My post was to show the difference between PR of two companies. MS circles around the word "eliminate load times" while Sony directly says "eliminate load times".

Honestly I just go with the demos that they showed us.

sony-presentation-wsj.jpg


sddefault.jpg


I'm sure that they will not miss the opportunity to show us better numbers.
 
Status
Not open for further replies.
Top Bottom