• Hey, guest user. Hope you're enjoying NeoGAF! Have you considered registering for an account? Come join us and add your take to the daily discourse.
  • The Politics forum has been nuked. Please do not bring political discussion to the rest of the site, or you will be removed. Thanks.

Xbox Series X’s BCPack Texture Compression Technique 'might be' better than the PS5’s Kraken

Mar 23, 2018
7,597
9,174
775
They showed on the fly asset loading in the spiderman demo.

Demo was also in a heavily controlled environment, which would make it work because they know at all times what's happening a second infront of them. That's the exactly bottleneck they walk against with that technique.

Btw. Michael ( NXG ) has been working as a software & hardware engineer for a long time. Surely he knows something. Better than rest of us here. Based on what Cerny himself said about only needing couple of percentage frequency clock drop when the system hits its set power limit, a tiny drop of which would claw back a huge amount of power (10%). Based on that, NXGamer has calculated around 50 MHz.

Cerny also needs to sell that box at the end of the day and tell you whatever u wanna hear and doesn't tell you what u don't wanna hear. Isn't it funny how we always hear about bottlenecks after a gen is over and not before? exactly.

If NXG gamer knows something he can share his sources of information if not, its just yet another voice in the crowd.
 
Last edited:

it_wasn't_me

Banned
May 22, 2017
1,991
3,129
395
Demo was also in a heavily controlled environment, which would make it work because they know at all times what's happening a second infront of them. That's the exactly bottleneck they walk against with that technique.

Like State Of Decay 2 then. And yet still in SOD 2 there were really visible texture loading on XSX and that was official showcase, but Spiderman was under development
 
Last edited:

it_wasn't_me

Banned
May 22, 2017
1,991
3,129
395
Never saw state of decay 2, link? and what did they mention?


And yet still in SOD 2 there were really visible texture loading on XSX and that was official showcase, but Spiderman was under development like 10 months ago
 
Last edited:

darkinstinct

...lacks reading comprehension.
Jul 24, 2015
4,787
2,811
715
Switzerland
Like State Of Decay 2 then. And yet still in SOD 2 there were really visible texture loading on XSX and that was official showcase, but Spiderman was under development
Low effort troll. State of Decay 2 was the Xbox One game running on Series X, no optimizations. Spiderman was the heavily optimized version of the game running a very specific test scene.
 
Mar 23, 2018
7,597
9,174
775

And yet still in SOD 2 there were really visible texture loading on XSX and that was official showcase, but Spiderman was under development like 10 months ago

I dunno what you try to proof with this point? this game is straight up there to showcase load speed improvements? and does exactly that. Spiderman demo was created to showcase SSD streaming speed over HDD and load speeds.
 

it_wasn't_me

Banned
May 22, 2017
1,991
3,129
395
Low effort troll. State of Decay 2 was the Xbox One game running on Series X, no optimizations. Spiderman was the heavily optimized version of the game running a very specific test scene.

Like Spiderman is a PS4 game. Spiderman video is more than 10 months old and it was labeled as "under development", SOD 2 was labeled as OFFICIAL showcase recently. And also, both loading tests showed exactly same scene on X1 and XSX. Surely it is optimized for SSD loading.
 
Last edited:
  • Like
Reactions: Deto and ethomaz

it_wasn't_me

Banned
May 22, 2017
1,991
3,129
395
I dunno what you try to proof with this point? this game is straight up there to showcase load speed improvements? and does exactly that. Spiderman demo was created to showcase SSD streaming speed over HDD and load speeds.

Same as SOD 2, yet you see a texture loading on XSX. Did you see that on Spiderman? Mind you that both games are current gen games.
 
Last edited:
Mar 23, 2018
7,597
9,174
775
Same as SOD 2, yet you see a texture loading on XSX. Did you see that on Spiderman?

I don't think you understand what they try to showcase with both cases.

Xbox series x uses that game to showcase you the load speed differences between the HDD and SSD in a game that's builded for a HDD. spiderman demo was to showcase u how fast they could get the game to load and how a SSD would perform in fast motion over a HDD.
 
Last edited:

yurinka

Member
Jan 19, 2007
12,545
3,702
1,685
Barcelona, Spain
www.capcom-town.es
Looking at the SDD raw vs compressed speed specs in both consoles, we see that Kraken's compression seems to be better than Kraken: it boost the SDD raw loading times in a bigger percentage than in Sony's case.

But since Sony's SSD is way faster in raw data and has all the I/O tweaks, it compensates that, and the final result is that when using MS and Sony compression PS5 loads compressed data faster at more or less twice Xbox's speed: 4.8GB/s (MS) vs 8-9GB/s (Sony).
 
  • Like
Reactions: alabtrosMyster
Mar 23, 2018
7,597
9,174
775
=
Loading the game and assets loading in both SOD 2 and Spiderman.

State of decay showcased what the load speed improvements are for a current gen title, nothing else matters in that demo.
Spiderman demo showcased what u could do with a fast SSD in a controlled environment + load speed if optimized. Gameplay for example doesn't matter in the demo.

Both have different goals to showcase something. They are not the same.
 
Last edited:

Three

Member
Oct 26, 2014
6,171
3,727
610
Demo was also in a heavily controlled environment, which would make it work because they know at all times what's happening a second infront of them. That's the exactly bottleneck they walk against with that technique.
Please tell me what is a 'heavily controlled environment'. No that's not the case. They limit current gen player speed based on asset loading bottlenecks. They can now achieve faster player speeds without hitting that bottleneck
Check the GDC talk:
 
  • Like
Reactions: Matsuchezz

JägerSeNNA

Banned
Mar 23, 2020
47
57
110
Germany
If that were the case, they would just go with fixed clocks. Makes no sense to have a variable rate with such minor changes. Which means he is wrong.
That makes me also wonder why such a minor thing should be announced in front of public if you don’t have any reservations about it for the future🤔
 

it_wasn't_me

Banned
May 22, 2017
1,991
3,129
395
=


State of decay showcased what the load speed improvements are for a current gen title, nothing else matters in that demo.
Spiderman demo showcased what u could do with a fast SSD in a controlled enviroment + load speed if optimized. Gameplay for example doesn't matter in the demo.

Both have different goals to showcase something. They are not the same.

Spiderman is a current gen title. How it is optimized when it was labeled as "under development"

You have the part when game is loading and later fast traversal

 
Mar 23, 2018
7,597
9,174
775
Please tell me what is a 'heavily controlled environment'. No that's not the case. They limit current gen player speed based on asset loading bottlenecks. They can now achieve faster player speeds without hitting that bottleneck
Check the GDC talk:

He's talking about the PS4 game and its limitations and tricks how it works, i watched it for about 70% earlier on, so if he went in on PS5 part i have no idea about that part because i didn't watch it. Anyway dude made a good presentation that's for sure.

I am talking about the PS5 demo showcase that was a specific version that showcases the potential of a SSD and everything they showcased as far as i can remember in that video was in a controlled enviroment.

Spiderman is a current gen title. How it is optimized when it was labeled as "under development"

You have the part when game is loading and later fast traversal


What has under development to do with anything? the game was builded for the PS4. I am talking about the PS5 spiderman showcase.
 

RaySoft

Member
May 31, 2013
1,019
1,609
760
Oslo, Norway
What about the memory bus feeding the XSX GPU at a MUCH HIGHER bandwidth? That's 560 GB/s vs 448GB/s , a 112 GB/s advantage for the X feeding the memory pipeline vs a 22GB/s best case really vs 6GB/s, really 8-9 like you stated ,for the ps5. That's a Delta of 112GB/s for the X versus a likely Delta of 3GB/s( 9 - 6) for the ps5....much more the advantage for the series X. Not all apples to apples but still talking about the memory system setup and bandwidth with the same destinations. I don't think the xsx ssd will be as much as a bottleneck as the ps5's RAM bandwidth will be for it IMO.
Nice percentage calculation there champ!
 

it_wasn't_me

Banned
May 22, 2017
1,991
3,129
395
He's talking about the PS4 game and its limitations and tricks how it works, i watched it for about 70% earlier on, so if he went in on PS5 part i have no idea about that part because i didn't watch it. Anyway dude made a good presentation that's for sure.

I am talking about the PS5 demo showcase that was a specific version that showcases the potential of a SSD and everything they showcased as far as i can remember in that video was in a controlled enviroment.



What has under development to do with anything? the game was builded for the PS4. I am talking about the PS5 spiderman showcase.

I don't get it. That was a PS4 Spiderman on PS5. The same crap like it was SOD 2 for XSX
 
  • Like
Reactions: ethomaz

Three

Member
Oct 26, 2014
6,171
3,727
610
He's talking about the PS4 game and its limitations and tricks how it works, i watched it for about 70% earlier on, so if he went in on PS5 part i have no idea about that part because i didn't watch it. Anyway dude made a good presentation that's for sure.

I am talking about the PS5 demo showcase that was a specific version that showcases the potential of a SSD and everything they showcased as far as i can remember in that video was in a controlled enviroment.



What has under development to do with anything? the game was builded for the PS4. I am talking about the PS5 spiderman showcase.
You keep saying 'controlled environment' but never explain what you're implying even when asked directly to elaborate. what is controlled are you saying? The way the game is set up to stream assets would be the same on the PS5 but the fast drive allows loading that data much faster. What are you saying is different or 'controlled'?
 

Goliathy

Member
Mar 18, 2020
440
1,423
340

And yet still in SOD 2 there were really visible texture loading on XSX and that was official showcase, but Spiderman was under development like 10 months ago

Like Spiderman is a PS4 game. Spiderman video is more than 10 months old and it was labeled as "under development", SOD 2 was labeled as OFFICIAL showcase recently. And also, both loading tests showed exactly same scene on X1 and XSX. Surely it is optimized for SSD loading.



Demo uses backward compatible Xbox title to demonstrate load time technology and does not represent gameplay optimized for Xbox Series X.

So, this wasn’t Xbox enhanced. Just the regular Xbox one title. They just wanted to show the improved loading titles for ALL XBOX ONE titles WITHOUT developer input at all. Automatically. Using BC.

There was NOTHING optimized. The Spider-Man demo was just a demo. Specifically designed and developed for this demo. Not real world.
 
Last edited:

RaySoft

Member
May 31, 2013
1,019
1,609
760
Oslo, Norway
Looking at the SDD raw vs compressed speed specs in both consoles, we see that Kraken's compression seems to be better than Kraken: it boost the SDD raw loading times in a bigger percentage than in Sony's case.

But since Sony's SSD is way faster in raw data and has all the I/O tweaks, it compensates that, and the final result is that when using MS and Sony compression PS5 loads compressed data faster at more or less twice Xbox's speed: 4.8GB/s (MS) vs 8-9GB/s (Sony).
Like I said elsewhere.. That compressed read bandwidth is probably most-cases scenario, instead of a theoretical max. (that are usually listed in specsheets)
 
Last edited:

Rolling_Start

Banned
Jul 25, 2019
563
761
325
Like Spiderman is a PS4 game. Spiderman video is more than 10 months old and it was labeled as "under development", SOD 2 was labeled as OFFICIAL showcase recently. And also, both loading tests showed exactly same scene on X1 and XSX. Surely it is optimized for SSD loading.

SoD2 isn't optimised for DirectStorage, and won't be using anything like BCPack. It's probably optimised for access patterns on a mechanical HDD. It's a stock X1 game.

Spiderman demo is definitely not just the standard PS4 game, as it allows you to move much faster than the actual game caps your movement.

So Spiderman does have some changes, but we don't know the extent. As an SSD demo it would make sense to use it show off the SSD.

Same as SOD 2, yet you see a texture loading on XSX. Did you see that on Spiderman? Mind you that both games are current gen games.

But they aren't both current gen binaries running via BC. Only SoD2 is. Spiderman demo is PS5 demo compiled for the PS5.

Very, very different beasts.
 
Mar 23, 2018
7,597
9,174
775
You keep saying 'controlled environment' but never explain what you're implying even when asked directly to elaborate. what is controlled are you saying? The way the game is set up to stream assets would be the same on the PS5 but the fast drive allows loading that data much faster. What are you saying is different or 'controlled'?

I already mentioned it before i think, it means that the content is on rails/ controlled, they know what data comes and when and can plan for it which eliminates the SSD weakness.
 

Ascend

Member
Jul 23, 2018
3,538
5,158
585
The compression technique is better. It doesn't mean that it's faster. The numbers are clear.

The XSX has the capability to transfer 2.4GB/s of raw data. Compressed, it's 4.8GB/s, which means 100% more data.
The PS5 has the capability to transfer 5.5GB/s of raw data. Compressed, it's 9 GB/s. That's 64% more data.

The XSX compression is superior. But because the SSD speed is more limited in the XSX, the ultimately transfer speed will still be faster on the PS5. In fact, the PS5 can still send more data per second without any compression, than what the XSX can with its max compression.
 

it_wasn't_me

Banned
May 22, 2017
1,991
3,129
395
SoD2 isn't optimised for DirectStorage, and won't be using anything like BCPack. It's probably optimised for access patterns on a mechanical HDD. It's a stock X1 game.

Spiderman demo is definitely not just the standard PS4 game, as it allows you to move much faster than the actual game caps your movement.

So Spiderman does have some changes, but we don't know the extent. As an SSD demo it would make sense to use it show off the SSD.



But they aren't both current gen binaries running via BC. Only SoD2 is. Spiderman demo is PS5 demo compiled for the PS5.

Very, very different beasts.

It's not. Because it was also tested on PS4
 

Rolling_Start

Banned
Jul 25, 2019
563
761
325
It's not. Because it was also tested on PS4

Yes, PS4 Pro was compared to a version running on a very early devkit. That doesn't mean that it was the PS4 binary running on the PS5 kit devkit.

It was not a BC demo. It was an SSD demo. I doubt BC was even fully nailed in April last year (infact it still isn't). Just putting the PS4 game on an SSD, even if the PS4 binary could take full advantage of the Zen 2, wouldn't allow this kind of speedup as there are bottlenecks elsewhere. To quote Cerny himself:

The raw read speed is important,“ Cerny says, “but so are the details of the I/O [input-output] mechanisms and the software stack that we put on top of them. I got a PlayStation 4 Pro and then I put in a SSD that cost as much as the PlayStation 4 Pro—it might be one-third faster." As opposed to 19 times faster for the next-gen console, judging from the fast-travel demo. "

That's what he demoing. Not BC mode.
 

yurinka

Member
Jan 19, 2007
12,545
3,702
1,685
Barcelona, Spain
www.capcom-town.es
Like I said elsewhere.. That compressed read bandwidth is probably most-cases scenario, instead of a theoretical max. (that are usually listed in specsheets)
Yes, Cerny said that these 8-9GB/s were more or less the average, common scenario. That if particularly well compressed it could go as high as 22GB/s.

But the idea was the same: PS5 puts in a second around 2x the amount of data in the memory after being copied and decompressed from the SSD than Series X. The lead is clear.

We also see when comparing raw vs compressed in both consoles that MS has a small lead in compression. But it's tiny when looking at the final results of also considering the SSD raw+I/O speed.
 
  • Like
Reactions: it_wasn't_me

RaySoft

Member
May 31, 2013
1,019
1,609
760
Oslo, Norway
Yes, Cerny said that these 8-9GB/s were more or less the average, common scenario. That if particularly well compressed it could go as high as 22GB/s.

But the idea was the same: PS5 puts in a second around 2x the amount of data in the memory after being copied and decompressed from the SSD than Series X. The lead is clear.

We also see when comparing raw vs compressed in both consoles that MS has a small lead in compression. But it's tiny when looking at the final results of also considering the SSD raw+I/O speed.
Yes, but do they rly have a small lead in compression though? Sony's numbers are real-life rounded numbers, while what I read from the other camp is more teoretical numbers (usually max numbers) so the comparison is probably more like apples to bananas...
 
Last edited:
  • Like
Reactions: Neo_game
Jan 29, 2019
6,070
6,594
495
State of decay showcased what the load speed improvements are for a current gen title, nothing else matters in that demo.
It sure loads faster, especially if it doesn't load completely.

Granted the Spiderman demo shows something different, it's dynamic loading instead of static loading, both show different benefits of faster storage media and potential use cases... As you said, it's not really something you can compare directly.
 

LittleBusters

Banned
Jun 4, 2018
2,422
2,037
465
42
Yes, but do they rly have a small lead in compression though? Sony's numbers are real-life rounded numbers, while what I read from the other camp is more teoretical numbers (usually max numbers) so the comparison is probably more like apples to bananas...

It's interesting that you choose to believe Sony and not MIcrosoft. Very interesting. Wow.
 

RaySoft

Member
May 31, 2013
1,019
1,609
760
Oslo, Norway
It's interesting that you choose to believe Sony and not MIcrosoft. Very interesting. Wow.
We already know that Cerny's numbers are more real-life numbers (average), since devs have come come out and stated they get much more than the stated 8-9GB/s.
Then you have MS side where "they" calculate with BCPack doing 8:1 compression ratio, wich it's max.. Do you think it can sustain 8:1 on every single texture?
A more apples to apples comperison then is to use max ratio's for both compression libraries, would'nt you agree?
 
Last edited:

South

Banned
Feb 13, 2020
91
126
230

And yet still in SOD 2 there were really visible texture loading on XSX and that was official showcase, but Spiderman was under development like 10 months ago

So about 10 times faster ?
So if the rate is meant to be from HDD to SDD a factor of 100 or 200 then we only see real world performance of 10x.


To be honest loading from 1min to 16 seconds is not draw dropping, IF the PS5 could do the same around 8 seconds then that would be amazing.
Saying this though also shows real world if the ps5 is 2x the series x ssd then the real world difference will be around 10-20%
 

Jigsaah

Gold Member
Jan 31, 2018
5,952
6,090
740
You guys have so much stamina. I'm starting to feel like I'm not nearly excited enough for next gen. I just want to buy both and don't care which one is better in what way.
 

INCUBASE

Member
Jan 8, 2018
6,890
13,817
730
Can't wait to unpack all the games, like........errm......err.......halo? And others I'm sure
 

Rolling_Start

Banned
Jul 25, 2019
563
761
325
We already know that Cerny's numbers are more real-life numbers (average), since devs have come come out and stated they get much more than the stated 8-9GB/s.

I think Cerny was trying to give a realistic, general case range for compression knowing that it varies significantly based on use but that some uses will be more common than others.

Then you have MS side where "they" calculate with BCPack doing 8:1 compression ratio, wich it's max.. Do you think it can sustain 8:1 on every single texture?

I haven't seen MS claim 8:1 compression on anything already compressed, let alone every single (already compressed) texture.

It's easy to get high (lossy) compression ratios on uncompressed textures.

Highly compressing textures in such a way that only fractions of mip maps can be read and decompressed - with high efficiency - is where the big gains for streaming are. This is where MS have been focusing their research. And it's fantastic, and a smart place to look at efficiency. But there is no indication it will outperform Sony's storage solution at this time. And I don't expect it to.

A more apples to apples comperison then is to use max ratio's for both compression libraries, would'nt you agree?

No, that's asinine.
 

RaySoft

Member
May 31, 2013
1,019
1,609
760
Oslo, Norway
I think Cerny was trying to give a realistic, general case range for compression knowing that it varies significantly based on use but that some uses will be more common than others.



I haven't seen MS claim 8:1 compression on anything already compressed, let alone every single (already compressed) texture.

It's easy to get high (lossy) compression ratios on uncompressed textures.

Highly compressing textures in such a way that only fractions of mip maps can be read and decompressed - with high efficiency - is where the big gains for streaming are. This is where MS have been focusing their research. And it's fantastic, and a smart place to look at efficiency. But there is no indication it will outperform Sony's storage solution at this time. And I don't expect it to.



No, that's asinine.
The 8:1 is the compression rayio on uncompressed textures (no need to compress twice) and I rly hope for their sake it's lossless, since it quite useless (as you point out) otherwise.

My point on comparing compression was that the compairing should be equal. Always better to compare real-world numbers ofc, but then the others should too.
 

Rolling_Start

Banned
Jul 25, 2019
563
761
325
My point on comparing compression was that the compairing should be equal. Always better to compare real-world numbers ofc, but then the others should too.

Yep, real world use cases are where you really find out what works well and what doesn't.

There's still a lot to be seen there and it's early days. The one thing we can say, I think, with confidence is that there's not going to be any area of SSD access where the PS5 is weak.

Fast, and robust, seems to be the order of business.
 
  • Like
Reactions: RaySoft

Three

Member
Oct 26, 2014
6,171
3,727
610
I already mentioned it before i think, it means that the content is on rails/ controlled, they know what data comes and when and can plan for it which eliminates the SSD weakness.
Which SSD weakness? The content is streamed live based on several tiles the player is currently looking at. The data required is mentioned in the presentation along with the hero speed. It's the PS4 game. If you do the calculation backwards for hero speed it is far beyond what was even shown in that demo.
 
Last edited:

yurinka

Member
Jan 19, 2007
12,545
3,702
1,685
Barcelona, Spain
www.capcom-town.es
Yes, but do they rly have a small lead in compression though? Sony's numbers are real-life rounded numbers, while what I read from the other camp is more teoretical numbers (usually max numbers) so the comparison is probably more like apples to bananas...
When you compress something, the % of size reduction depends on both the compressed files and the compression algorythm. As an example, an already compressed texture image when compressed (let's say a .JPG with its compression set to max for this example) with that would reduced in size way less than an uncompressed image (let's say a .BMP). And also changes how much something it's compressed between let's say a video or an image versus text files.

A game puts a good amount of different data to the memory, that changes from game to game. So it's pretty difficult to get a "max" or even "average" number of how much this is going to compress the game data. I assume that for these numbers they tested it with different cases similar to the ones they expect to run and did choose for the compressed amount more or less average of common cases scenarios. I assume that a best case scenario for some game may be just a bit higher than that.

A case like the one Cerny said about "up to 22GB/s if particularly well compressed" would be just the result from a test that they may have done by looking at the limits with a file type that is best compressed with that compression algorythm and that they rearranged it to be particularly well compressed. Just more a theorical case than a real world scenario to be seen in a game.

Cerny mentions up to 22GB/s from 5.5GB/s raw data in his best case scenario, which is 4:1. MS claims 8:1 in his best case scenario, but their source raw speed is twice as slow, so the results in both cases would be pretty much the same. And these are best case theorical scenarios. The numbers they show in their specs would be more common case real world scenarios, and there we can see a small lead in compression in MS side, but since raw is speed is also taken in consideration for the final results, PS5 ends being around twice as fast as Series X on average for real world, practical cases.
 
Last edited:
  • Like
Reactions: Ascend

RaySoft

Member
May 31, 2013
1,019
1,609
760
Oslo, Norway
When you compress something, the % of size reduction depends on both the compressed files and the compression algorythm. As an example, an already compressed texture image when compressed (let's say a .JPG with its compression set to max for this example) with that would reduced in size way less than an uncompressed image (let's say a .BMP). And also changes how much something it's compressed between let's say a video or an image versus text files.

A game puts a good amount of different data to the memory, that changes from game to game. So it's pretty difficult to get a "max" or even "average" number of how much this is going to compress the game data. I assume that for these numbers they tested it with different cases similar to the ones they expect to run and did choose for the compressed amount more or less average of common cases scenarios. I assume that a best case scenario for some game may be just a bit higher than that.

A case like the one Cerny said about "up to 22GB/s if particularly well compressed" would be just the result from a test that they may have done by looking at the limits with a file type that is best compressed with that compression algorythm and that they rearranged it to be particularly well compressed. Just more a theorical case than a real world scenario to be seen in a game.

Cerny mentions up to 22GB/s from 5.5GB/s raw data in his best case scenario, which is 4:1. MS claims 8:1 in his best case scenario, but their source raw speed is twice as slow, so the results in both cases would be pretty much the same. And these are best case theorical scenarios. The numbers they show in their specs would be more common case real world scenarios, and there we can see a small lead in compression in MS side, but since raw is speed is also taken in consideration for the final results, PS5 ends being around twice as fast as Series X on average for real world, practical cases.
That 8:1 ratio must be either an absolute max peak which, in theory, will never be reached or it's not lossless compression. I find it hard to believe that someone suddenly invents an algorithm that's twice as effecient, than other competeable algorithms. when we've only seen incremental improvements for years. If it's not lossless at 8:1 it's ofcource doable, but in that case, you can't compare lossy compression against lossless. (hence my apples to bananas reference earlier)
We need more info to know for sure.
 

ethomaz

is mad because DF didn't do a video on a video of a video of a video on PS5
Mar 19, 2013
40,226
40,778
1,310
39
Brazil
It's interesting that you choose to believe Sony and not MIcrosoft. Very interesting. Wow.
It is not him.
Some devs are skeptical with MS numbers because it shows 50% compression that is really hard to believe for lossless compression for games... so they believe it is not the average, typical.

While Sony numbers are 30-40% that is more believable to be the average for loss less compression.
 
Last edited:
  • Like
Reactions: Deto

ethomaz

is mad because DF didn't do a video on a video of a video of a video on PS5
Mar 19, 2013
40,226
40,778
1,310
39
Brazil
When you compress something, the % of size reduction depends on both the compressed files and the compression algorythm. As an example, an already compressed texture image when compressed (let's say a .JPG with its compression set to max for this example) with that would reduced in size way less than an uncompressed image (let's say a .BMP). And also changes how much something it's compressed between let's say a video or an image versus text files.

A game puts a good amount of different data to the memory, that changes from game to game. So it's pretty difficult to get a "max" or even "average" number of how much this is going to compress the game data. I assume that for these numbers they tested it with different cases similar to the ones they expect to run and did choose for the compressed amount more or less average of common cases scenarios. I assume that a best case scenario for some game may be just a bit higher than that.

A case like the one Cerny said about "up to 22GB/s if particularly well compressed" would be just the result from a test that they may have done by looking at the limits with a file type that is best compressed with that compression algorythm and that they rearranged it to be particularly well compressed. Just more a theorical case than a real world scenario to be seen in a game.

Cerny mentions up to 22GB/s from 5.5GB/s raw data in his best case scenario, which is 4:1. MS claims 8:1 in his best case scenario, but their source raw speed is twice as slow, so the results in both cases would be pretty much the same. And these are best case theorical scenarios. The numbers they show in their specs would be more common case real world scenarios, and there we can see a small lead in compression in MS side, but since raw is speed is also taken in consideration for the final results, PS5 ends being around twice as fast as Series X on average for real world, practical cases.
8:1 is probably not not lossless compression.
Cerny 22GB/s is probably some data similar to text files that can reach 80% of lossless compression.
It is something that rarely will happen in games 90% of game data cases.
 
Last edited:

SLB1904

Member
Oct 24, 2017
2,677
5,214
575
doesn't really matter is better or not. doesn't mean it comes close to ps5 speeds. secret sauce all over again
 

SLB1904

Member
Oct 24, 2017
2,677
5,214
575
So about 10 times faster ?
So if the rate is meant to be from HDD to SDD a factor of 100 or 200 then we only see real world performance of 10x.


To be honest loading from 1min to 16 seconds is not draw dropping, IF the PS5 could do the same around 8 seconds then that would be amazing.
Saying this though also shows real world if the ps5 is 2x the series x ssd then the real world difference will be around 10-20%
ugh. ffs
 

Kokoloko85

Member
Sep 26, 2019
3,316
4,139
395
Assumptions and more assumptions just to make anything playstation says does look obsolete.
I wonder how much MS pays for its anti playstation marketing lol.....
 

phil_t98

Member
Oct 10, 2014
4,571
4,704
620
Assumptions and more assumptions just to make anything playstation says does look obsolete.
I wonder how much MS pays for its anti playstation marketing lol.....
why are they paying for marketing? same could be said of both sides.
 
  • Like
Reactions: Closer

svbarnard

Member
Mar 17, 2016
175
166
410