A Few 9700 Screenshots

[RussSchultz]

My point is, no one can predict what is going to happen.

You're right - which is why you're own argument is pretty pointless. All there is is the present, and at present ATI have executed and the competition is apparently months behind; EOA. So, whats your point?

If you notice, his argument wasn't NVIDIA r0x0rs. His argument was "declaring winners now is silly because who knows what the future holds" and came up with a completely valid alternative future that contradicts the previous persons prediction based on no less speculative information.

But yes, who knows what the future holds. I don't care, as long as its cheap DX9 parts for me.

 

[LeStoffer]

<speculation mode on>
The NV30 seems to be designed around the idea that there will be a paradigm shift to increasing computation per pixel, away from raw pixel fillrate. (e.g. bandwidth/fillrate now "good enough"). NVidia may be eyeing the workstation/renderfarm market as a side, but in general, they think the future is in shader performance.

...

</speculation mode off>

Yes, this is where we gotta be heading sooner or later, IMO. Shaders will be the primary rendering norm in the future and thus computational power will be key to fast performance. The more [complex] ops you can do per cycle [per shader], the better. I think that we will see specs in the future based on just this: How many cycles will a chip take to make this or that shader ops (read shader program/effect) and we should see some benchmarks that pinpoint this more clearly than just report avg framerates (which we of course still need).

This also suggest that we doesn't yet have much clue to how powerful Radeon 9700 or NV30 really is when we're talking about complex shaders. Is this just a first generation with most of the features but not the ops per cycle (OPC!) power to drive the programmabilty? Or are they actually fairly fast already?

At least we have got a lot of goodies with DX9 that we can built upon in the next two or three years. Lets delay the launch of DX10 and instead focus on making the future chips really, really fast for DX9. 8)

That is my focus - not whether NV30 will beat R9700 and if under what circumstances.

 

[Bambers]

I bought a geforceSDR with an athlon 500MHz in nov99 and I don't regret the purchase at all. The card lasted me 2 years playing games smoothly at 1024x768x32 (I only have a 15inch monitor but that doesnt mean I'm limited to 800x600, I can run 1024x768@85Hz and 1152x864 @ 80Hz), the tnt2 was only ahead in low resolutions/16bit and that was fixed very quickly after the release of the card.

Those pics look very nice, pity I can't afford a new gfx card at the moment.

 

[Joe DeFuria]

If you notice, his argument wasn't NVIDIA r0x0rs. His argument was "declaring winners now is silly because who knows what the future holds"

Problem is, and this seems to be a trend as of late, he is arguing against things no one is claiming. Nobody is "declaring winners" (ATI or nVidia) based on the R300 / NV30 situation. Yes, "declaring now" is indeed silly. That's why no one is doing it.

Here's the argument in sum:

Person A): Ati made it to market first, that's a good advantage to ATI, and the decision to use "proven and predictably available tech" had a direct role in ATI enjoying the R300 "first launch" victory.

Person B): How can you predict the destruction of nVidia based on this one product launch?

Person A): huh?

 

[Doomtrooper]

If you notice, his argument wasn't NVIDIA r0x0rs. His argument was "declaring winners now is silly because who knows what the future holds"

As stated many times before..the Radeon 9700 is not competing with the Nv30..do you think Nvidia is the only company working on .13, to think ATI will not answer Nv30..almost 5 months away...be it a highly overclocked 9700 with DDR II..who KNOWS

I just don't understand comments like this..especially someone supposedly familiar with the business.

 

[RussSchultz]

If you notice, his argument wasn't NVIDIA r0x0rs. His argument was "declaring winners now is silly because who knows what the future holds"

As stated many times before..the Radeon 9700 is not competing with the Nv30..do you think Nvidia is the only company working on .13, to think ATI will not answer Nv30..almost 5 months away...be it a highly overclocked 9700 with DDR II..who KNOWS

I just don't understand comments like this..especially someone supposedly familiar with the business.

Now its my turn to say "WTF?" I don't see the logical following.

Person A says "choice is right"
Person B says "we'll see how it works in the long run as we can't see the whole picture now"
Person C says "of course we can't see the whole picture, but choice is right because that's what we can see now"
Person D (me) says "ummm, you missed his point"
Person E says "nobody is saying choice is right"
Person F (you) comes in with a non-sequitor and a personal attack.

So, WTF? I never mentioned anybody being able to do anything better than anybody else, I never mentioned any product competing with any other product, one company's superiority, or what they're working on or what process they're working on, or when anything will come out.

So, just where did you pull your comments from?

 

[Joe DeFuria]

Russ,

How about this:

1) Clearly, it is evident at this time, that ATI's decision to base their first DX9 product on 0.15u and "predictably available memory", has given ATI the competitive advantage of getting their product to market first.

2) nVidia's decision to rely on 0.13 and presumably, "not so predictably available" memory, has put them at a disadvantage at this time, as they are late shipping.

3) Wheteher or not nVidia's decision to go with 0.13 gives them a "better" product, when it does ship, than ATI's at 0.15, remains to be seen.

4) In this market where the "high-end" products tend to have a practical shipping life of only 6 months to 1 year before they are discontinued, a few months difference in shipping time can make a very big difference in terms of the "success" of that product in terms of profitability.

 

[Doomtrooper]

You stated nobody is declaring a winner right now..I am ..ATI won and has time to tweak, die shrink ..whatever for Q1 2003.

Fastest card you can buy from Nvidia is a Ti4600, fastest card you buy from ATI is a Radeon 9700..

Its not that hard to see the winner here...some people just won't admit it...they keep using the 'but XXX is coming' so is a Opteron and possibly we even get visited by Aliens. There is always something better coming, the 9700 overclocking thread is classic...denial on a grand scale.
Instead of applauding for great engineering effort on .15 there is three pages of 'that can't be true'

 

[Joe DeFuria]

You stated nobody is declaring a winner right now..I am ..ATI won and has time to tweak, die shrink ..whatever for Q1 2003.

Well, you have to define what "winner" means. I certainly agree that ATI holds the advantage going forward. That now, nVidia is the one who has catching up to do if ATI continues to execute on its recent pace.

Edit: your definition of goes only as far as the next product goes. So if Nv30 comes out and is faster than R300, that would make nVidia the "winner", until ATI could counter that....seems like a rather meaningless definition of "winner" to me.

ATI has lived up to their "significant new core" every 9-12 months claim they made when the Radeon 8500 shipped. If ATI can follow up the R300 in less than a year with a significant new chip (not just faster clocked 9700), they will start to make nVidia's "acclaimed" execution schedule look poor in comparison.

 

[Doomtrooper]

I'm talking winner in terms of:

1) They executed and delivered
2) Pricing although high, is about the same as a Ti4600
3) Performance
4) Visual Quality is improved

They will have drivers working well by the Xmas season as they will have had 4-5 months to get maximum peformance out of their card...which is good for us.

 

[RussSchultz]

So, the weather here in Austin is nice. Matter of fact, the whole summer has been, except for the mosquitos. How's the weather in other parts?

 

[Joe DeFuria]

How can you claim that the weather in Austin is better than it's been here in NJ?! The year isn't even over yet, so it's sillly to start laying the groundwork for Austin being the winner...

 

[Sabastian]

My point is, no one can predict what is going to happen.

You're right - which is why you're own argument is pretty pointless. All there is is the present, and at present ATI have executed and the competition is apparently months behind; EOA. So, whats your point?

If you notice, his argument wasn't NVIDIA r0x0rs. His argument was "declaring winners now is silly because who knows what the future holds" and came up with a completely valid alternative future that contradicts the previous persons prediction based on no less speculative information.

But yes, who knows what the future holds. I don't care, as long as its cheap DX9 parts for me.

Excuse me, I think it is pretty clear at this point who is the winner. ATI has beat nvidia to the market by a clear margin for the first DX9 part. The Radeon 9700 clearly hammers Nvidias offerings.

The nv30 is some time in the future... who knows when you will be able to get one. This is a silly argument. ATI clearly holds the performance/technology leadership presently to suggest otherwise is foolhardy and dismissive of what ATI has managed with the Radeon 9700.

There simply is no reason to make a suggestion that ATI has not won... based on what the future will bring. If that was the case well... there never would be a leader ... ever. This sort of conclusion would fly in the face of those who have absolutely declared that nvidia was in the past "undisputed leader" in the graphics arena.

To put my argument as simply as possible you cannot declare or dismiss a technology leader based on what may/could/should happen in the future. If that were the case then no one would ever be a leader ever even if they truely are. Just my 2 cents.

 

[Doomtrooper]

Edit: your definition of goes only as far as the next product goes. So if Nv30 comes out and is faster than R300, that would make nVidia the "winner", until ATI could counter that....seems like a rather meaningless definition of "winner" to me.

I'm not talking about next generation at all ?? I'm talking this generation right now...ATI has plans for the January and DDR II is part of it.
All I'm saying is you were going to buy a card right NOW which one would you buy:
Ti4600
R9700

What card will give you the best visuals with the best speed and Direct 9 Compliancy ??

 

[martrox]

Hey...the weather in Florida really sucks......too much sun.......too many thong wearing babes......

 

[tamattack]

... NVidia has NV31/35/38/Mobile chips with significantly ramped up clocks and low power consumption...

Historically, NV has not been able to come close to matching ATI's low power consumption which is why ATI is still the leader in discrete mobile. Given NV's inability to deliver low power in the past, I find it a bit of a stretch to even speculate that NV will be able to deliver both ramped clocks and low power.

 

[John Reynolds]

I'm talking winner in terms of:

1) They executed and delivered
2) Pricing although high, is about the same as a Ti4600
3) Performance
4) Visual Quality is improved

Pricewatch shows GF4 Ti4600s starting around $240 USD. MSRP has dropped to around $330, at least at my local Best Buy.

 

[BenSkywalker]

2) Pricing although high, is about the same as a Ti4600

Do you have any links? I was just looking at a few Ti4600s for under $250, if a DX9 board can be had for under $300 then I'm certainly interested.

 

[Doomtrooper]

John I wish Canada has Pricewatch

Here its quite different:

Best price for the 9700 is:$583
http://www.oxycom.com/canada/final.asp?8249272xxqfg418=45913822&RECNO=61853

While Quality Ti4600's are going for:$598.00
http://www.ncix.com/canada/productd...mp;PHPSESSID=74db6a847b2010a30ff987784fc95b56

 

[Sabastian]

2) Pricing although high, is about the same as a Ti4600

Do you have any links? I was just looking at a few Ti4600s for under $250, if a DX9 board can be had for under $300 then I'm certainly interested.

Yeah they ought to be priced around midstream level considering how well the Radeon 9700 performs. I would suspect the prices have a bottom on these cards though. If the third party manufacturers start having to sell them at a loss to clear their inventories as a result of the Radeon 9500 card being introduced at or around the same price levels of the Geforce ti series cards then I would suspect we could see more of the same sorts of problems for nvidia third parties a la VisionTech. This would result in huge portions of market share being lost to ATI.