***OFFICIAL*** FX5700/FX5950 Thread

[Nebor]

Originally posted by: GTaudiophile
Just to prove my point again that Anand's conclusion is seriously flawed, someone over at Beyond3D tabulated these numbers based on Anand's review:

NON AA
---------
5700 wins 10 times
9600 XT wins 6

Where the 5700 won, it won on average by 15%
Where the 9600 won, it won on average by 17%


WITH AA / ANISO
---------
5700 wins 6 times
9600 wins 6 times

Where the 5700 won, it won on average by 23%
Where the 9600 won, it won on average by 54%

See, I wouldn't have a problem with this, except I know that if we went through and switched "5700" with "9600" above, you wouldn't be in for that fight. You're a fanatic.

 

[GTaudiophile]

I don't feel too bad Nebor. Anand's review is the only one I've read that has much such an outlandish conclusion.

 

[Nebor]

Originally posted by: GTaudiophile
I don't feel too bad Nebor. Anand's review is the only one I've read that has much such an outlandish conclusion.

You should let him know you think so. After all, you've been at this a lot longer than him. And your review was a lot better. What was the URL again?

 

[Sunner]

To me, personally, the 5700U seems like the winner.
But Im basing that very much on my oppinion that AA is the most overhyped feature since...don't know really, forever?

When I got my 9800 I looked forwards to seeing this gorgeous AA/AF everyone kept raving about, but in reality, I don't see even the slightest difference unless I stand still and have a closer look.
This is of course talking about games where I can simply raise the res, in older games where I can't set the res to 1600x1200(Freespace 1&2 for example) it makes a big difference.

Anyways, for that very reason, the 5700U would be an easy choice for me.
Depends on what you value the most, maybe Anand(Or rather Derek as Ulric pointed out) thinks the AA/AF numbers should'n count as high, just as I do?

 

[UlricT]

guys... the article was written by Derek, not Anand. I suppose the conclusion was written by him as well. So you can hardly blame anand for that.

Having said that, I seriously think that every review produced by Anandtech should go through some kind of Quality Control process. Otherwise, in time, AT will begin producing junk reviews like th one at Tom's.

 

[Chumster]

Originally posted by: Nebor

You should let him know you think so. After all, you've been at this a lot longer than him. And your review was a lot better. What was the URL again?

 

[GTaudiophile]

FYI, the EVGA FX5950 256 VIVO Retail is in stock @ newegg for $499.

ATi Radeon 9800XT 256 DVI/TV Retail in stock @ $539

ATi Radeon 9800XT 256 DVI/TV OEM in stock @ $495

 

[jm0ris0n]

Originally posted by: GTaudiophile
FYI, the EVGA FX5950 256 VIVO Retail is in stock @ newegg for $499.

ATi Radeon 9800XT 256 DVI/TV Retail in stock @ $539

ATi Radeon 9800XT 256 DVI/TV OEM in stock @ $495

Whats your point? The 5950 is $40 cheaper and has vivo functionality ?

 

[GTaudiophile]

The point really was, if you got $500 to burn on a videocard, they are available for purchase.

 

[Nebor]

Still waiting for you to link to your Anand stompin' review, GTAudiophile.....

 

[GTaudiophile]

Originally posted by: Nebor
Still waiting for you to link to your Anand stompin' review, GTAudiophile.....

Here you go, Nebor!

 

[Lyfer]

Originally posted by: Sunner
To me, personally, the 5700U seems like the winner.
But Im basing that very much on my oppinion that AA is the most overhyped feature since...don't know really, forever?

When I got my 9800 I looked forwards to seeing this gorgeous AA/AF everyone kept raving about, but in reality, I don't see even the slightest difference unless I stand still and have a closer look.
This is of course talking about games where I can simply raise the res, in older games where I can't set the res to 1600x1200(Freespace 1&2 for example) it makes a big difference.

Anyways, for that very reason, the 5700U would be an easy choice for me.
Depends on what you value the most, maybe Anand(Or rather Derek as Ulric pointed out) thinks the AA/AF numbers should'n count as high, just as I do?

Its pretty much even with the 9600 XT, but it cost $220 and comes with no Halflife 2. I think the 9600 Pro is a MUCH better value of the two. If you're gonna fork up $200, why not spend a little more and get a 9800NP/9700Pro.

 

[Corporate Thug]

Originally posted by: Nebor

Originally posted by: GTaudiophile
I don't feel too bad Nebor. Anand's review is the only one I've read that has much such an outlandish conclusion.

You should let him know you think so. After all, you've been at this a lot longer than him. And your review was a lot better. What was the URL again?

just because he does it longer doesnt mean that he doesnt make mistakes like the rest of us

 

[Medicated858]

Originally posted by: GTaudiophile

Originally posted by: Nebor
Still waiting for you to link to your Anand stompin' review, GTAudiophile.....

Here you go, Nebor!

That review rocked! The fx5700 looks like a Barney of a card. As your review stated both companies seem quite childish.

 

[ElFenix]

ati seriously needs to up their memory clock rate.

 

[JavaMomma]

Originally posted by: Lyfer

Originally posted by: Sunner
To me, personally, the 5700U seems like the winner.
But Im basing that very much on my oppinion that AA is the most overhyped feature since...don't know really, forever?

When I got my 9800 I looked forwards to seeing this gorgeous AA/AF everyone kept raving about, but in reality, I don't see even the slightest difference unless I stand still and have a closer look.
This is of course talking about games where I can simply raise the res, in older games where I can't set the res to 1600x1200(Freespace 1&2 for example) it makes a big difference.

Anyways, for that very reason, the 5700U would be an easy choice for me.
Depends on what you value the most, maybe Anand(Or rather Derek as Ulric pointed out) thinks the AA/AF numbers should'n count as high, just as I do?

Its pretty much even with the 9600 XT, but it cost $220 and comes with no Halflife 2. I think the 9600 Pro is a MUCH better value of the two. If you're gonna fork up $200, why not spend a little more and get a 9800NP/9700Pro.

 

[Corporate Thug]

Originally posted by: ElFenix
ati seriously needs to up their memory clock rate.

not necessarily, their cards scale better. its almost like the amd/intel comparison

 

[Sunner]

Originally posted by: Lyfer

Originally posted by: Sunner
To me, personally, the 5700U seems like the winner.
But Im basing that very much on my oppinion that AA is the most overhyped feature since...don't know really, forever?

When I got my 9800 I looked forwards to seeing this gorgeous AA/AF everyone kept raving about, but in reality, I don't see even the slightest difference unless I stand still and have a closer look.
This is of course talking about games where I can simply raise the res, in older games where I can't set the res to 1600x1200(Freespace 1&2 for example) it makes a big difference.

Anyways, for that very reason, the 5700U would be an easy choice for me.
Depends on what you value the most, maybe Anand(Or rather Derek as Ulric pointed out) thinks the AA/AF numbers should'n count as high, just as I do?

Its pretty much even with the 9600 XT, but it cost $220 and comes with no Halflife 2. I think the 9600 Pro is a MUCH better value of the two. If you're gonna fork up $200, why not spend a little more and get a 9800NP/9700Pro.

Actually, I forked out $300 for a 9800

Anyways, my point was, that since I don't care about AA, which is where the 9600 has it's stronghold, the 5700 would make more sense to me.

 

[Blastman]

NVIDIA has flipped the tables on ATI in the midrange segment and takes the performance crown with a late round TKO.

I?d hardly call that review a knockout for the 5700U. The cards were very close overall. Going by the numbers --- (adding up Anand?s numbers) ? 16 benchmarks (1024x768) ? 5700U ? had only a 3.3% average advantage over 9600XT , 13.9% over the 9600pro. Only 12 benches were provided at 1024x768 4AA/8AF --- 9600XT had a 15.4% average advantage over the 5700. 9600pro was 2.8% slower than 5700.

That being said, the 5700U is a huge improvement over the 5600U. The 5600U had some serious performance problems in a lot of games and the 5700U and drivers seem to have fixed that.

THG?s numbers ? ?

9 benchmarks (1024x768) ? 9600XT had a 6.4% advantage over 5700U --- 9600pro was 3.5% slower than 5700U. At higher resolutions (1600x1200) the 5700U turns back the tide in its favor and outruns the 9600XT. THG?s inclusion of DX9 benches like Tomb Raider helps the 9600XT and pulls things in it?s favor. The 5700U looks to be perhaps a little better in older titles and at higher resolutions. The 9600XT is still looking to be a lot better in newer tiles (DX9).

With the 9600pro covering the $150 market ATI should have put better memory on the 9600XT to justify the $200 price tag and to keep the midrange lead. With 450MHz DDR (like the 5700U) the 9600XT could have easily beaten the 5700U. As it is the 9600XT is not really a significant jump over the 9600pro. ATI really needed to fill the gap between a $150 -- 9600pro ... and a $250 -- 9800 so the 9600XT should have had at least 400MHz DDR. But it seems ATI wants to save money (make money) and the 9600XT is competitive with the 5700U anyway as it is. A faster 9600XT would have justified the $200 price (and premium over the 9600pro.) and since the 9600XT is really no more expensive to make than the 9600pro it could come down in price to the 9600pro?s level fairly quickly, and then this would leave a gap around the $200 mark again.

 

[GTaudiophile]

Review from Driver Heaven

Despite the apparent raw speed from the specifications however I can?t help but think back to a conversation I had with a member of the graphics card industry recently, the comment that sticks in my mind was that as we advance in the industry memory bandwidth is becoming less important. The most important factor now is the core of the card and how well it performs in processing items like shaders. From our tests it is clear to see this is becoming more and more the case and the Radeon provides a far more efficient architecture for today?s and tomorrows titles.

So who should consider this card as an upgrade? Put up against the 9600 XT I just can?t recommend anyone to buy the 5700 Ultra. It?s clear that Nvidia can squeeze performance out of the card?3DMark shows that. However people buy cards to play games and there is no way Nvidia can optimise their FX range for performance in every game that is released. Nothing emphasises the situation FX users are going to be in more often than not than Max Payne 2, a game that is a week old and which no manufacturer has had time to optimise for, there really is no competition in the performance here.

 

[Regs]

Are there any benchmarks on Max Payne 2 Ati vs Nvidia? If the ATi cards, stock, kick the crap out of Nvidia in Max Payne 2, it would completely change the way I think of which graphic cards to choose from or to recommend.

There should be no excuse for Nvidia's FX not to be able to keep up with the ATi in Max Payne 2, no excuse.

That's all I've been hearing as of late, is excuses.

 

[GTaudiophile]

Originally posted by: Regs
Are there any benchmarks on Max Payne 2 Ati vs Nvidia? If the ATi cards, stock, kick the crap out of Nvidia in Max Payne 2, it would completely change the way I think of which graphic cards to choose from or to recommend.

There should be no excuse for Nvidia's FX not to be able to keep up with the ATi in Max Payne 2, no excuse.

That's all I've been hearing as of late, is excuses.

ATi vs. nVidia in Max Payne 2

 

[InlineFive]

nVidia's cards just keep getting bigger and more cumbersome with every new model.

 

[GTaudiophile]

Originally posted by: PorBleemo
nVidia's cards just keep getting bigger and more cumbersome with every new model.

I also find it interesting that the FX5700 still requires a secondary power connector, whereas the 9600XT does not. But it's probably better to be safe than sorry and include one for overclocking purposes.

 

[maluckey]

I have .02 cnets for this one. Derek was astute to notice that the high framerates skew the results. In real-time play, the absolute low, and average FPS is far more important than the abslute high rate. Derek clearly stated that the 5700 Ultra could hold substantially higher minimum framerates in some of he tests where it scored lower than the ATI products. This speaks volumes for the fairness of the test. How most benchmarks work is on average FPS, not sustained average V. absolute minimum. The fact that he pointed this out shows that he does know what he is talking about, and all those that only read benchmarks are missing the true picture. The 9700 Pro is a fster card, and there is not a shred of doubt. The mid-range is pretty clear, in that the 5700 Ultra can maintain higher average, and absolute low FPS than the other mid-range offerings. Look at the big picture, not a few tests, and a scoresheet.

Nobody seems to believe synthetic benches that speak against their favorite hardware, but embrace them fully when they are in agreement with them. Many people read a benchmark without considering the other factors, that are possibly equally important. I for one wouldn't hesitate to buy a 5700Ultra, but if I had the money, I'd buy the 9700 Pro, because there is no doubt at all about it's capabilities, and it is faster than the 5700 Ultra (as it should be).