"PS3 & Xbox360 in the Real World"

Clearly written by a fan and not a developer. Too many basic mistakes, and too much misleading info to be a real developer.
 
By the way, I heard at one point that 772.6 is greater than both 278.4 and 48 combined. That proves that Nintendo Revolution will be the most powerful console ever!....

There you go, Revolution will be the bestestest ever.

Nice article clear target market for this kind of article - the above average, but not quite mathematically programming genius (like faf or ERP), interweb console fan. Not many people will listen to this guy but I liked it.
 
london-boy said:
I like my eyes too much to read the whole thing. I'm sorry that's just painful, anyone reading that whole thing without getting a brain tumor gets major man points.
I agree that the font is kinda' small, and the white text against dark blue background is kinda' annoying :D ..... perhaps copy/paste in WordPad would be a good idea? :D
 
Last edited by a moderator:
Powderkeg said:
Clearly written by a fan and not a developer. Too many basic mistakes, and too much misleading info to be a real developer.

Hey well, feel free to point out the aspects you disagree with Powderkeg.

I personally thought it was well written, and as a counterbalance to MajorNelsons blog and Anandtech's article serves it's purpose well. Besides when you say 'a fan not a developer,' that's as if to say that it spares one console over the other. They both certainly enjoy their fair share of cuts in that piece, and I've heard a number of Sony fans complain that it's 'too favorably 360 biased.' So, basically people see what they want to see.

EDIT: Tahir 2, I thought that Revolution bit was pretty hilarious also. ;)
 
Last edited by a moderator:
Oh by the way it was written by Shootmymonkey. It is a quite well written piece in my opinion.
 
xbdestroya said:
Hey well, feel free to point out the aspects you disagree with Powderkeg.

A couple of examples of basic errors....

The GPU of Xbox360 has a daughter die containing 10 MB of DRAM plus a little bit of extra logic. The bandwidth between this logic and the DRAM is 256 GB/sec. However, this die cannot communicate with other devices at this rate or anywhere near it. The bandwidth between this daughter die and the functional part of the GPU is only 32 GB/sec. The bandwidth between the GPU and any other device is at most 22.4 GB/sec. No matter how fast the eDRAM is, it will never be able to transfer to and from other devices faster than that 22.4 GB/sec.

While that is technically correct, it is misleading in that the bandwidth of the eDRAM that is being used is bandwidth that would normally be taken from the Main System Memory. So, while you shouldn't add the two together (He is correct on that) you should keep in mind that the bandwidth that the eDRAM provides is saving bandwidth across the main memory bus.


Measuring point-to-point memory bandwidth only is actually more meaningful anyway since it’s this bandwidth that has always been a limiting factor for performance. Bandwidth within the silicon area of a single chip is rarely a problem until a chip becomes too complex for its own good. By that measure, Xbox360 has at most 22.4 GB/sec to PS3’s 48 GB/sec.

This is severely flawed logic, and contridicts his own claims. If you measure point to point bandwidth, then the PS3 most certainly does not have 48GB/sec bandwidth. Cell cannot access the GDDR-3 RAM, so there is no point-to-point communication there. Basically, he just did precisely what he just got done saying you shouldn't do. That is, he added the bandwidth of two unrelated memory banks that do two seperate jobs in the system.

If you were going to accurately measure point to point bandwidth, then RSX has 22.4GB/Sec to work with, and Cell has 25GB/Sec. RSX can access the XDR Memory, but that takes both RAM and bandwidth away from cell, and also introduces some rather disturbing latency issues.

The whole piece is packed with little errors like that. It's pretty clear that it's a gamers attempt at appearing "unbiased" and informed, but not the comments of an actual developer.
 
Last edited by a moderator:
I read the whole thing. Can't say I was impressed. It looks like he skimmed a bunch of B3d posts, took them out of context and formed an article out of it. For example, regarding memory latency, he cites forum posts talking about 500+ cycle latency for the xbox. If I recall correctly, this discussion was actually about both systems.

Cell cannot access the GDDR-3 RAM

I was under the impression that it can.

While that is technically correct, it is misleading in that the bandwidth of the eDRAM that is being used is bandwidth that would normally be taken from the Main System Memory. So, while you shouldn't add the two together (He is correct on that) you should keep in mind that the bandwidth that the eDRAM provides is saving bandwidth across the main memory bus.

I agree. It isn't nearly as irrelevant as the article suggests.
 
Last edited by a moderator:
Powderkeg said:
While that is technically correct, it is misleading in that the bandwidth of the eDRAM that is being used is bandwidth that would normally be taken from the Main System Memory. So, while you shouldn't add the two together (He is correct on that) you should keep in mind that the bandwidth that the eDRAM provides is saving bandwidth across the main memory bus.

I don't think he says otherwise - it's just not addressed, but yes I agree that the eDRAM would relieve the main memory bandwidth to serve other functions.

This is severely flawed logic, and contridicts his own claims. If you measure point to point bandwidth, then the PS3 most certainly does not have 48GB/sec bandwidth. Cell cannot access the GDDR-3 RAM, so there is no point-to-point communication there.

If you were going to accurately measure point to point bandwidth, then RSX has 22.4GB/Sec to work with, and Cell has 25GB/Sec. RSX can access the XDR Memory, but that takes both RAM and bandwidth away from cell, and also introduces to rather disturbing latency issues.

Well, I don't agree that it's 'severely flawed,' it is still system bandwidth in a sense, but I see where you're coming from. Still - I don't think the statement is completely specious.

Basically, he just did precisely what he just got done saying you shouldn't do. That is, he added the bandwidth of two unrelated memory banks that do two seperate jobs in the system.

The whole piece is packed with little errors like that. It's pretty clear that it's a gamers attempt at appearing "unbiased" and informed, but not the comments of an actual developer.

Yet, he *is* a dev, so I think your 'obvious' take on it comes up short in that sense. ;)

But anyway - try reading past the memory bandwidth paragraph. Obviously that was there as a Major Nelson retort and nothing more.
 
xbdestroya said:
Yet, he *is* a dev, so I think your 'obvious' take on it comes up short in that sense. ;)

A PC developer at last check. He made it abundently clear during the Xenos FSAA debate that he has little more than a cursory understanding of the Xenos architecture and he states clearly in the article that he isn't a PS3 developer. Not to mention his embarassing attempt at bashing the xbox(1) API. He has been called out by actual console developers, like ERP, on several occasions for his inaccuracies.
 
Last edited by a moderator:
So all the article really says:

1)It looks like both consoles will have similiar performance
2)Nobody in the public really knows how hard they will be to program, especially compared to each other
3)They will have inferior single threaded and logic/data performance to a P4 or XP
4)They won't be as good as the ******s think, especially to begin with.

Admittedley it did just sort of become a blure towards the end.
 
SanGreal said:
A PC developer at last check. He made it abundently clear during the Xenos FSAA debate that he has little more than a cursory understanding of the Xenos architecture and he states clearly in the article that he isn't a PS3 developer. Not to mention his embarassing attempt at bashing the xbox(1) API. He has been called out by actual console developers, like ERP, on several occasions for his inaccuracies.

SanGreal I really do not want to get into this with you. I'm in what I feel is an unusual situation for myself since not only am I a staff member for the site from whence the article came, I'm friends with the author as well.

I think this article should be viewed as what it was meant to be: a counterbalance to Anandtech and MajorNelson posted on a Playstation site. I certainly feel it is better informed and less inflamatory than either of those two were, respectively. Not meant as a treatise on the matter, to be sure.

On the issue of 'real' devs and what not, well - he 'really' is a console dev. The fact that up until recently he was a PC dev doesn't change what he is now. If you like, he's a rookie in that field - but it is his field. I give ERP, nAo, DeanoC and the rest of the veteren programming crew here as much credit as any other true B3D member; but not being a dev myself I'm not in a position to get involved in these developer 'feats of strength.'

Anyway like I said this is not the usual topic for me here, so please bare with my obtuse responses.
 
Fair enough, there is no reason to bash the author of the article, it is only the content which is important. I disagree that the article was lacking in bias and inflamatory remarks, but perhaps that is just my own bias speaking.
 
london-boy said:
I like my eyes too much to read the whole thing. I'm sorry that's just painful, anyone reading that whole thing without getting a brain tumor gets major man points.

ctrl+scroll wheel down is a wonderful thing. ;)
 
Powderkeg said:
Cell cannot access the GDDR-3 RAM
Of course it can, don't be rediculous.

If you were going to accurately measure point to point bandwidth, then RSX has 22.4GB/Sec to work with, and Cell has 25GB/Sec.
The two chips also share 20GB/s write and 15GB/s read performance (seen from perspective of cell) buses between the two of them. I don't see why this bandwidth should be ignored when clearly it is required to take advantage of both main memory and video memory pools to reach maximum utilization of system resources.

and also introduces some rather disturbing latency issues.
What do you know about "disturbing"? This is completely your own fabrication.

You might as well say PC graphics cards have "disturbing" latency when accessing main RAM in AMD64 systems too, since they have to go through the CPU's memory controller in that case also. In practice though, AMD64 systems usually lead in games performance compared to systems that use the traditional architecture of a memory controller on the same chip as the graphics bus controller, so obviously there's not neccessarily any inherent problems with this design setup.

If you wish to claim the situation is otherwise for PS3, you better show some cold hard facts to back up your position...

The whole piece is packed with little errors like that.
Perhaps. Or perhaps it's just your own misconceptions and misinterpretations that made you think it is. :devilish:
 
SanGreal said:
Fair enough, there is no reason to bash the author of the article, it is only the content which is important. I disagree that the article was lacking in bias and inflamatory remarks, but perhaps that is just my own bias speaking.

I could see where you would perceive bias against the MS side of things; I don't see things as being as biased in the article as you seem to, but then again that may be my own bias effecting things.

I'll certainly accept the olive branch offer however. :)
 
Last edited by a moderator:
Nice article clear target market for this kind of article - the above average, but not quite mathematically programming genius (like faf or ERP), interweb console fan.
Yep, that's basically the typical visitor of the site. A B3D or Ars crowd would really get nothing out of it, which is fine -- I didn't write it for B3D or Ars. Also, when originally asked to do the piece, the biggest thing they wanted was for me to refute Major Nelson's nonsense. I kind of felt Major Nelson's blog piece was water under the bridge, but I figured if it even briefly silenced the seemingly neverending "kill Major Nelson" and "laugh at TeamXbox" threads, it's worth it.

I guess we still need something to counter all the "IGN are idiots" threads.

My only biggest fear was that I might have let something slip out that shouldn't have been said. So I played it cautious in keeping everything down to the publically available knowledge, a lot of which, happens not to be true.

I agree that the font is kinda' small, and the white text against dark blue background is kinda' annoying
Ummm... there should be a link for a "print" layout that will give you 12pt black text on a white background.

And yeah, I don't care for that layout anymore than anyone else on that site does.
 
so whats the conclusion in the real world then? that it will be up to the devs? and both systems will be within spitting distance of each other?
 
Back
Top