G70 Vs. R520 All things being equal

[jiffylube1024]

Originally posted by: keysplayr2003
I wonder if the extra 512MB of memory will help the G70 with it's AA/AF stamina.

I'm curious as well. My speculation has been that the 512MB on the X1800XT is pretty much all hype (just like the 256MB 9800 Pros), and that the main reason for the X1800XT's high res performance is it's super fast RAM (1.5 Ghz!).

We'll see if/when the new Nvidia 512MB cards come out just how big of an impact the memory makes...

 

[1Dark1Sharigan1]

Originally posted by: LTC8K6
So, what will ATI have if they get 20 or 24 "pixel processors" going considering they are doing pretty well with just 16?

The R580 . . .

Originally posted by: jiffylube1024
I wonder if the extra 512MB of memory will help the G70 with it's AA/AF stamina.

[/quote]

It's been said before but I think memory bandwidth (speed) is probably more important when it comes to AA/AF than framebuffer size . . .

 

[Hacp]

How will they compare 4 pipes 100 MHZ core 200 MHZ ram????? I wonder....

 

[reever]

So, what does this article prove, does it affect anything in ANY way? Is anything new learned from this article?

 

[CaiNaM]

Originally posted by: reever
So, what does this article prove, does it affect anything in ANY way? Is anything new learned from this article?

that some people have far too much time on their hands?

why does an orange taste like an orange, but an apple taste like an apple?

 

[kruull]

Originally posted by: jiffylube1024

Originally posted by: Acanthus
That article has no bearing on anything. Im not even going to get into it.

Exactly. While the 'efficiency' debate is interesting, it's nonetheless irrelevant. R520 doesn't run at 450 MHz and you'd be a fool to run your 7800GTX @ 16 pipelines.

People buy Athlon64's because they are faster, not because they are more efficient than Prescott P4's.

Just to be correct - these were 1800XL and 7800GT!!! Look the standard clocks of the R520 tested and the pipelines of G70...

 

[imported_Shivetya]

Originally posted by: Acanthus
That article has no bearing on anything. Im not even going to get into it.

Actually it does. It shows that once Nvidia releases their .90 proc'd GPUs coupled with the same memory as ATI that their cards will beat ATI's R520s. Worse for ATI is the fact that Nvidia will not have to change their archietecture to do it.

This isn't great news for gamers as it means the R520s are not good enough to keep pushing Nvidia forward. Clearly the R520s are one to two quarters late in release. I would go as far as saying that the R520s are probably not worth upgrading from the previous generation of ATI cards and in fact the Nvidia G70 series is a better value.

Hopefully ATI can work some driver magic otherwise there is nothing to see here.

 

[Keysplayr]

Originally posted by: kruull

Originally posted by: jiffylube1024

Originally posted by: Acanthus
That article has no bearing on anything. Im not even going to get into it.

Exactly. While the 'efficiency' debate is interesting, it's nonetheless irrelevant. R520 doesn't run at 450 MHz and you'd be a fool to run your 7800GTX @ 16 pipelines.

People buy Athlon64's because they are faster, not because they are more efficient than Prescott P4's.

Just to be correct - these were 1800XL and 7800GT!!! Look the standard clocks of the R520 tested and the pipelines of G70...

The test wasn't comparing a X1800XL to a 7800GT. It was comparing an R520 to a G70.

And just so you know, the 7800GT has only 7 vetex units. (1 disabled). Model numbers don't mean much anymore when you change frequencies/pipes like that.

 

[Keysplayr]

Originally posted by: Shivetya

Originally posted by: Acanthus
That article has no bearing on anything. Im not even going to get into it.

Actually it does. It shows that once Nvidia releases their .90 proc'd GPUs coupled with the same memory as ATI that their cards will beat ATI's R520s. Worse for ATI is the fact that Nvidia will not have to change their archietecture to do it.

This isn't great news for gamers as it means the R520s are not good enough to keep pushing Nvidia forward. Clearly the R520s are one to two quarters late in release. I would go as far as saying that the R520s are probably not worth upgrading from the previous generation of ATI cards and in fact the Nvidia G70 series is a better value.

Hopefully ATI can work some driver magic otherwise there is nothing to see here.

Have to completely disagree with you there. Thats not what the test shows at all. Its a no brainer that if Nvidia delivered a 90nm G70 with 512MB of 1500MHz memory and the core bumped up to higher clocks that it would handily beat out a current spec'd X1800XT. This test was to compare architecture efficiency only.

 

[Keysplayr]

Originally posted by: reever
So, what does this article prove, does it affect anything in ANY way? Is anything new learned from this article?

You tell us. I got what I needed out of it. Did you?

 

[CaiNaM]

Originally posted by: keysplayr2003

Originally posted by: reever
So, what does this article prove, does it affect anything in ANY way? Is anything new learned from this article?

You tell us. I got what I needed out of it. Did you?

what did you need to get out of it?

 

[Keysplayr]

Originally posted by: CaiNaM

Originally posted by: keysplayr2003

Originally posted by: reever
So, what does this article prove, does it affect anything in ANY way? Is anything new learned from this article?

You tell us. I got what I needed out of it. Did you?

what did you need to get out of it?

Well, not a "need" really. Just a curiosity satisfaction is all. Like I mentioned earlier, I often wondered how similarly clocked competitors would do against each other and often wondered who had the more effective arch. I had an X800XTPE AGP and a 6800GT AGP a while back. The 6800GT ran 370/1000 and the PE ran 520/1120. I downclocked the PE to the GT speeds, and well, you know what the outcome of that was.

 

[CaiNaM]

ya.. sure do.

but strickly from a technical aspect i still don't see anything useful here, other than perhaps it shows the r3xx architecture (and yes, other than the revamped memory architecture and some other tweaks, it's still more r3xx core than not) has pretty much reached it's limits, which explains why ati will go a completely different architecture here very shortly. course, we didn't really need this comparison to tell us that..

kind of like intel -- they could only go so far with minor core changes and ramping clockspeeds until they finally hit that "wall" with their technology...

 

[Munky]

Originally posted by: Shivetya

Originally posted by: Acanthus
That article has no bearing on anything. Im not even going to get into it.

Actually it does. It shows that once Nvidia releases their .90 proc'd GPUs coupled with the same memory as ATI that their cards will beat ATI's R520s. Worse for ATI is the fact that Nvidia will not have to change their archietecture to do it.

This isn't great news for gamers as it means the R520s are not good enough to keep pushing Nvidia forward. Clearly the R520s are one to two quarters late in release. I would go as far as saying that the R520s are probably not worth upgrading from the previous generation of ATI cards and in fact the Nvidia G70 series is a better value.

Hopefully ATI can work some driver magic otherwise there is nothing to see here.

Actually, I disagree., for several reasons:

1) The r520 has a more efficient memory controller, and thus even if Nv used the same memory speed, it would still suffer a bigger hit from AA+AF. Look at this page: http://www.extremetech.com/article2/0,1697,1867129,00.asp and compare the overall performance of the 7800gt vs. the x1800xl. They both use 1 ghz mem, and while the xl is overall slower without AA, they're even with AA, so it suffers less of a hit.

2) The r520 is designed in such a way that it does a lot better when running shaders with branching than the gtx. Look at the second chart here http://www.techreport.com/reviews/2005q4/radeon-x1000/index.x?pg=14. The Nv cards actually suffer a hit when runing shaders with flow control (branching), while the r520 gets a big boost. That means when a game is using complex SM3 shaders with branching, the 7800 will do WORSE than if it were running the same shader without branching, which kinda kills the whole point of branching. I have a feeling that's the reason why it loses pretty bad in modern games like BF2 and FEAR demo. To combat this problem, the developers will have to restructure and reorganize their code to make it run better on the 7800, or NV might even go back to shader replacement.

3) The r520 drivers are nowhere near maturity, and I expect another 10% or more gain in performance in the near future with better drivers.

 

[Keysplayr]

Originally posted by: CaiNaM
ya.. sure do.

but strickly from a technical aspect i still don't see anything useful here, other than perhaps it shows the r3xx architecture (and yes, other than the revamped memory architecture and some other tweaks, it's still more r3xx core than not) has pretty much reached it's limits, which explains why ati will go a completely different architecture here very shortly. course, we didn't really need this comparison to tell us that..

kind of like intel -- they could only go so far with minor core changes and ramping clockspeeds until they finally hit that "wall" with their technology...

Arghghh.... Yes, Intel is a whole other agonizing story. Hopefully they will get their act together soon enough.

I agree that this in no way should reflect which GPU to buy. This is only for "fun" so to speak. I have emailed the author of that article at DriverHeaven and asked if he would be so kind to add a few top 10 game benches in there. HL2, FarCry, SC:CT, CoD2, FEAR, UT2004, etc. I doubt it will happen, but at least I asked.

 

[Munky]

Originally posted by: CaiNaM
ya.. sure do.

but strickly from a technical aspect i still don't see anything useful here, other than perhaps it shows the r3xx architecture (and yes, other than the revamped memory architecture and some other tweaks, it's still more r3xx core than not) has pretty much reached it's limits, which explains why ati will go a completely different architecture here very shortly. course, we didn't really need this comparison to tell us that..

kind of like intel -- they could only go so far with minor core changes and ramping clockspeeds until they finally hit that "wall" with their technology...

Did you read anything about the r520 archtecture? Obviously not, because the g70 is alot more similar to the nv40 than the r520 is to the r300. If you really want to claim it's the same, then read this http://www.beyond3d.com/reviews/ati/r520/ and tell me if you stil think it's the same thing. In fact, the r520 will be the basis of the r580, which will emphasize shader power, and not just slapping on more pipes. Because they separated the texture units from the pixel shaders, it breaks the whole concept of a "pipe", and is in fact half-way there on the road to unified shaders.

 

[Keysplayr]

Originally posted by: munky

Originally posted by: CaiNaM
ya.. sure do.

but strickly from a technical aspect i still don't see anything useful here, other than perhaps it shows the r3xx architecture (and yes, other than the revamped memory architecture and some other tweaks, it's still more r3xx core than not) has pretty much reached it's limits, which explains why ati will go a completely different architecture here very shortly. course, we didn't really need this comparison to tell us that..

kind of like intel -- they could only go so far with minor core changes and ramping clockspeeds until they finally hit that "wall" with their technology...

Did you read anything about the r520 archtecture? Obviously not, because the g70 is alot more similar to the nv40 than the r520 is to the r300. If you really want to claim it's the same, then read this http://www.beyond3d.com/reviews/ati/r520/ and tell me if you stil think it's the same thing. In fact, the r520 will be the basis of the r580, which will emphasize shader power, and not just slapping on more pipes. Because they separated the texture units from the pixel shaders, it breaks the whole concept of a "pipe", and is in fact half-way there on the road to unified shaders.

Munky, you're starting to sound like a PR spinner. This is a "for fun" thread if anything. Try to keep it that way.

 

[Munky]

Originally posted by: keysplayr2003

Originally posted by: munky

Originally posted by: CaiNaM
ya.. sure do.

but strickly from a technical aspect i still don't see anything useful here, other than perhaps it shows the r3xx architecture (and yes, other than the revamped memory architecture and some other tweaks, it's still more r3xx core than not) has pretty much reached it's limits, which explains why ati will go a completely different architecture here very shortly. course, we didn't really need this comparison to tell us that..

kind of like intel -- they could only go so far with minor core changes and ramping clockspeeds until they finally hit that "wall" with their technology...

Did you read anything about the r520 archtecture? Obviously not, because the g70 is alot more similar to the nv40 than the r520 is to the r300. If you really want to claim it's the same, then read this http://www.beyond3d.com/reviews/ati/r520/ and tell me if you still think it's the same thing. In fact, the r520 will be the basis of the r580, which will emphasize shader power, and not just slapping on more pipes. Because they separated the texture units from the pixel shaders, it breaks the whole concept of a "pipe", and is in fact half-way there on the road to unified shaders.

Munky, you're starting to sound like a PR spinner. This is a "for fun" thread if anything. Try to keep it that way.

It's all fun and games until someone actually believes the post I replied to.

 

[classy]

Comparisons like this are so stupid. They have absolutely ZERO relevance and offer ZERO insight.

 

[Keysplayr]

Originally posted by: classy
Comparisons like this are so stupid. They have absolutely ZERO relevance and offer ZERO insight.

Stupid to some. Fun for others. Get the proverbial grip.

 

[ArchAngel777]

Originally posted by: classy
Comparisons like this are so stupid. They have absolutely ZERO relevance and offer ZERO insight.

I am not so sure about that... I think the comparison was cool. I enjoyed it.

 

[rise]

Originally posted by: ArchAngel777

Originally posted by: classy
Comparisons like this are so stupid. They have absolutely ZERO relevance and offer ZERO insight.

I am not so sure about that... I think the comparison was cool. I enjoyed it.

people try to make too much out of it or get very defensive for some reason

 

[classy]

Originally posted by: keysplayr2003

Originally posted by: classy
Comparisons like this are so stupid. They have absolutely ZERO relevance and offer ZERO insight.

Stupid to some. Fun for others. Get the proverbial grip.

Tests like this are totally irelevant, because they are two completely different architectures. Any conclusions would be based on assumption.

 

[CaiNaM]

Originally posted by: munky

Originally posted by: CaiNaM
ya.. sure do.

but strickly from a technical aspect i still don't see anything useful here, other than perhaps it shows the r3xx architecture (and yes, other than the revamped memory architecture and some other tweaks, it's still more r3xx core than not) has pretty much reached it's limits, which explains why ati will go a completely different architecture here very shortly. course, we didn't really need this comparison to tell us that..

kind of like intel -- they could only go so far with minor core changes and ramping clockspeeds until they finally hit that "wall" with their technology...

Did you read anything about the r520 archtecture? Obviously not, because the g70 is alot more similar to the nv40 than the r520 is to the r300. If you really want to claim it's the same, then read this http://www.beyond3d.com/reviews/ati/r520/ and tell me if you stil think it's the same thing. In fact, the r520 will be the basis of the r580, which will emphasize shader power, and not just slapping on more pipes. Because they separated the texture units from the pixel shaders, it breaks the whole concept of a "pipe", and is in fact half-way there on the road to unified shaders.

lol.. did you?

i'll stick with what i said: r520 carries over most of the design principles of r420, and r300 before it. ati did revamp the memory architecture (as i acknowledged), however the vertex shaders and pixel shaders all follow the principles of r420 and earlier (the obviously had to extend it's capabilites to (almost fully) support sm3.

while the updated memory architecture did increase some of the efficiency (and granted, this is a bit of an oversimplification as there are other areas ati has attempted to optimize with varying degress of success, but that's beyond the scope of this dicussion), it was nowhere near enough of a design change to keep ati competetive. moving to 90nm was a must so ati could ramp up the clockspeeds (way beyond r420's) in order to keep pace.

so yea, all in all r520 is an evolution of the desing pricnciples of r420 and r300 before that. it may have been intentional (to smooth the move to 90nm) or they may have been forced to as they simply weren't ready to apply other techniques.. either way, more radical changes are in the forefrong as they just can't keep looking at raising clockrates (as they have the past 3 generations) to raise performance.

 

[ArchAngel777]

Originally posted by: rise4310

Originally posted by: ArchAngel777

Originally posted by: classy
Comparisons like this are so stupid. They have absolutely ZERO relevance and offer ZERO insight.

I am not so sure about that... I think the comparison was cool. I enjoyed it.

people try to make too much out of it or get very defensive for some reason

? Having a hard time finding the angle you are coming from. Anyway, I think not everything has to be ATI Versus nVidia. This was a very good read, just interesting to see is all. However, I do find the tests relevent for the following reasons.

1. Generally clock speeds can only go so fast. If ATI was able to get 600Mhz core from 90nm tech, then I doubt nVidia's chip would have much of a problem being clocked that high either, in a 90nm setting. Of course they are still on 110nm, so that can be debated.

2. Memory clock speeds are an issue cross platform. If nVidia has access to 1.xx ns ram chips, then ATI has that option as well. If one can be more effecient with memory, cheaper costs being able to get by with slower memory clocks, or better performance. Either way, it is a win-win situation to have a more effecient design.

3. I am aware these tests are not the end-all-be-all, they are just an interesting read.

No need to get worked up about it, I mean, it is only tech talk, right?