Devs Speak on PS3.0

Hellbinder said:
GraphixViolence said:
I particularly liked how Andrey from CryTek gave his little checklist of all the supposedly useful new features of SM3.0... including 16-bit FP render targets and MRT which the 9700 Pro has already had for the last year and a half :rolleyes:
LOL, Ati has been orffering FP24 Render targets... Even better FP24 *MULTIPLE* Render Targets since the R300's Launch :)

He was talking about FP16 blending num-nut. (multi-pass HDR rendering)
 
Mark Davis said:
What will really clinch it is the 'killer app' that requires PS3.0 in order to look its best. If Doom3 adds PS3.0 support then I'll bet that that's enough for gamers to want to buy into the new technology in big numbers.

Yeah alright carmack is gonna change from openGl to D3D to support the new PS 3.0.

Mark Davis said:
Besides, if card X runs Q3 at 170fps and card Y runs Q3 at 180fps, who cares? As long as I get at least 60fps, I'd rather have prettier graphics.

How about we use a morden game to look at the FPS say like far cry rather then a game that doesn't use any sort of DX8.0+ gen pixel shader? since no game is out that supports PS3.0 wouldn't it mean that current games are wrong to measure the expected FPS of future games that gonna us more complex effects?

Most of the other comments made I thought where valid but I dunno about this.

Tim Sweeny said:
(such as position vector math, world transformations, etc - all things one is likely to do frequently per-pixel),

Correct me if I'm wrong but isn't this done Per Vertex????
 
I found some comments startlingly facetious and marketing-laden. :-?

Tim Sweeny said:
...
How many useful C programs don't contain conditionals? Approximately none!
...
Only a marketing guy would consider 24-bit floating point to be "full precision"!
...
But nowadays there isn't a good reason for hardware to support less than full 32-bit floating point precision.

I found this an unfortunate inclusion.

In the rest of his comments, the usage of words like "obvious" and "huge" give the entire thing the sound of an advertisement, though the actual things discussed in that other commentary seem otherwise informative. Sort of matching the comments by Rowan that are on the "opposite side" :-?, with a similar touch of hype and focus on their own applications specifically, except that Rowan comments of that type happened to talk less expansively

OTOH, I found Andrey's and Rick's comments the most thoughtful and generally informative about the actual question. There was some turning of a blind eye to whether some things NV40 offered with its "PS 3.0/SM 3.0/CineFX 3.0" that it is trying to make synonymous, could actually be (or have already been done) without PS 3.0 on another card by Andrey, but that seems reasonably interpretd to be the result of the context of the question.

I found Bart's response the most gamer-centric and unadorned, while making it clear that it was restricted to his experience and opinion.
 
Keep in mind, that the ones who are more positive of it happen to be part of NVIDIA?s ?The Way it?s Meant to be Played? Program, while a couple that are less supportive are with ATI?s ?Get in the Game? Program.
Am I the only one to find that rather pathetic? If we can't trust developers to be independent, who can we trust? Sad.
 
demalion said:
I found Bart's response the most gamer-centric and unadorned, while making it clear that it was restricted to his experience and opinion.

It suggests to me that he's probably doing the least forward looking development right now.
 
A good reason why I returned UT 2004, developers part of the TWIMTBP are worse than a used car saleseman. Ever seen the movie 'Used Cars' with Kurt Russell, that is exactly how I picture Sweeny. :LOL:

It used to be developers coded for the API, and used the strengths of all cards, now its 'show me the money' and I'll make your graphic card shine.
 
ninelven said:
Well, so far on NV4X, fp16 has only been shown to be a performance win in lighting situations as opposed to the NV3X hardware. I do expect we will be encountering it, but I would think less instead of more. I guess it could eventually become an issue with registers maybe?
As far as I know, the only thing that's faster with FP16 on NV40 is the NRM macro (for vector normalization).
As an external storage format it does make a difference, too, but for the shader units itself it's supposed to not matter anymore. *ahem*
 
Doomtrooper said:
A good reason why I returned UT 2004, developers part of the TWIMTBP are worse than a used car saleseman.
You returned a brilliant game because you don't like the lead engine programmers views? :?

Reading the article it does seem that the developers most keen on SM3.0 are the ones that are developing the most cutting edge engines. When it comes down to engine programming I trust sweeney/carmack/crytek over some guy who works for Gas or wrote the engine for Tribes.
 
DaveBaumann said:
demalion said:
I found Bart's response the most gamer-centric and unadorned, while making it clear that it was restricted to his experience and opinion.

It suggests to me that he's probably doing the least forward looking development right now.

:?: Actually, Rowan's comments struck me as the least forward looking, because they emphasized a limited subset of even PS 2.0.

Bart simply talked about not seeing things beyond PS 2.0 as being "must have" and focused on PS 2.0 and minimum requirements, which would indeed be quite significantly forward looking in combination.

What he was the "least" at, IMO, was being talkative about the possibilities of technology and what it could achieve (as in pretty much not at all), and therefore leaving the reader the most uninformed about what his development was. I called this unadorned because this happened to be all that was required to answer the question and it made no pretensions on speaking further than his own perspective in recognition of what it lacked.

It is, of course, without the interesting technical discussions and evaluations of some other comments, so might be the "most boring" as well. :p
 
Bart said:
We like to ensure our games work on the majority of video hardware regardless. The card vendor certification initiatives seem to affect marketing more than development.

Thank you Bart! That's what I wanted to hear in that article :D

Mark said:
Besides, if card X runs Q3 at 170fps and card Y runs Q3 at 180fps, who cares? As long as I get at least 60fps, I'd rather have prettier graphics.

Wow what a great example man. Really who cares about Q3 and its image quality -- not like you can improve it in anyway *sigh* :rolleyes:
 
Anyway, if you remeber reverend's posts from last year when UT 2003 cheats were discovered, he basically told he could not/would not be straight forward due to the financial deals involved with nvidia.
 
PatrickL said:
Anyway, if you remeber reverend's posts from last year when UT 2003 cheats were discovered, he basically told he could not/would not be straight forward due to the financial deals involved with nvidia.

yes after that comment i take tim sw. comments with alittle more grain of salt than before
 
It's really sad to see how developers cater to their financiers and their publishers when voicing an opinion on future features and the impact it has on consumers purchasing decisions.

Basically it reads like yes it’s important because our financier says so. No it’s not important because our financier says so.

They should have asked the questions like this, “if you didn’t get paid by X how important is PS 3.0 over 2.0 as pertains to CURRENT CARD purchasing and future proofing that purchase for lets say the next 18 monthsâ€￾

I would bet my house that the answers would have been quit different on all counts.

As it stands now I take all of their comments with a grain of salt, call me cynical, but I’m way to experienced to see it any other way.

As for Andrey Khonich – Crytek, Makers of Far Cry, all I have to say is get to work. Your game needs serious patching and a hell of a lot more support than you have been giving the people that paid for the game.
 
DW Fan!!!!! said:
Thats your opinion and I understand your point, the problem is, its actually amazing what TWIMTBP members will do to protect Nvidia (do you remember TRAOD).

Also almost all Devs will have been at those same events including those which were interviewed, these are people who are'nt fooled by Nvidia's propaganda.

I personally think BOTH Get In The Game and TWIMTBP are BAD things and we do not need them, they are purely for marketing and that is what they are doing but you do not agree and that is fine as everyone has there own point of view and opinion of whats going on around them.

I guess we will have to agree to disagree ;)

Its pretty much because Nivida is sending them a fat Check to get there logo on the game box. So of course they are going to protect Nivida.
 
Anybody who still believes that PS2.0 is better than PS3.0 should read this article. For a development standpoint, I get the feeling that PS3.0 is 100's of times better than PS2.0 because of the flow control and other parts like multiple light sources.

I didn't see anything wrong with Epic's response - unless you in ATI marketting and response that only marketing guys would like 24bit instead 32bit precision.

Just my input on this topic.
 
Back
Top