Yep, and it looks like I'm not the only one that picked up on this:
You'll find similar reactions and commentary without too much trouble, as most reviews I read picked up on the 4890's headroom as a pretty big deal.
Yep, and it looks like I'm not the only one that picked up on this:Originally posted by: SlowSpyder
Still saying this, are you... :roll:
You'll find similar reactions and commentary without too much trouble, as most reviews I read picked up on the 4890's headroom as a pretty big deal.Firing Squad
From ATI, the Radeon 4890 generally runs anywhere from 6-9% faster than the Radeon 4870 it displaces, but most importantly this card has a beefier cooler that addresses the concerns we?ve had with ATI?s previous stock cooler used on the 4870. The heat issues we?ve been nagging ATI about since launch have finally been resolved! Temps go from the 80 degree range to the 60s. ATI has also addressed the other shortcoming of RV770: frequency scaling.
RV790 scales like no other ATI GPU we?ve seen; when ATI told us that 1.0GHz was reachable with 4890, we thought they were referring to a card with exotic cooling or some one-off sample they?d scrounged up in the labs, but we now have no doubt that its achievable with the stock ATI cooler ? we didn?t even have to manually crank up the fan speeds in order to hit 1GHz either! In hindsight, it looks like ATI could have introduced an even more extreme 4890 SKU, but it appears they plan to leave that up to their board partners.
Originally posted by: zod96
I like how the 4890 is shorter than the gtx275 and how the power connectors are on the side rather than the top.
Originally posted by: josh6079
Originally posted by: zod96
I like how the 4890 is shorter than the gtx275 and how the power connectors are on the side rather than the top.
Actually, I'd rather ATi put the power connectors on the top like the GT200s do. Sure, the 4890 is shorter, but then you have to account for the room the PCI-E connectors are going to take and the slack needed. That actually extends the space required for it to fit in a smaller case.
Had they put the power connectors on the top, the shorter pcb of the 4890 would show more of an advantage than it does now. But, because they didn't, you'll have to make room at the end to allow the connections.
Originally posted by: chizow
Yep, and it looks like I'm not the only one that picked up on this:Originally posted by: SlowSpyder
Still saying this, are you... :roll:
You'll find similar reactions and commentary without too much trouble, as most reviews I read picked up on the 4890's headroom as a pretty big deal.Firing Squad
From ATI, the Radeon 4890 generally runs anywhere from 6-9% faster than the Radeon 4870 it displaces, but most importantly this card has a beefier cooler that addresses the concerns we?ve had with ATI?s previous stock cooler used on the 4870. The heat issues we?ve been nagging ATI about since launch have finally been resolved! Temps go from the 80 degree range to the 60s. ATI has also addressed the other shortcoming of RV770: frequency scaling.
RV790 scales like no other ATI GPU we?ve seen; when ATI told us that 1.0GHz was reachable with 4890, we thought they were referring to a card with exotic cooling or some one-off sample they?d scrounged up in the labs, but we now have no doubt that its achievable with the stock ATI cooler ? we didn?t even have to manually crank up the fan speeds in order to hit 1GHz either! In hindsight, it looks like ATI could have introduced an even more extreme 4890 SKU, but it appears they plan to leave that up to their board partners.
Originally posted by: SlowSpyder
My 2900 Pro overclocked 40%.
Originally posted by: error8
Originally posted by: SlowSpyder
My 2900 Pro overclocked 40%.
That doesn't count, since it was an underclocked XT all along. :laugh:
Originally posted by: chizow
AT benches? Is this a joke? The AT benches that show the GTX 275 leading by <1 FPS difference in all but a single benchmark where the GTX 275 also beats the GTX 285? I'd say <1 FPS difference is hardly conclusive. Not to mention the GTX 280 wins by much more than that where it pulls away.
Rofl, yes, the same benches where I had to drop bandwidth by 30% to see a 3% drop in performance. So what are you saying now? That bandwidth is more important than SP and TMU? You can't have it both ways. The fact the 275 does perform similarly to the 280 despite its decreased bandwidth proves bandwidth isn't the greatest bottleneck on GT200 parts.
No, it wouldn't, it'd show exactly how much difference those additional 4 ROPs yielded. The most accurate test would be clocking all the parts the same, GTX 260, 275, 280/285 and then seeing what differences yielded the biggest difference based on the actual hardware, not clockspeeds. Your 275 example doesn't tell us anything we didn't know already when we saw overclocked GTX 260s coming close to GTX 280 performance months and months ago. :laugh:
1. GTX280 | 2. GTX 275
ROP fillrate
19264 - 17724 = +8%
GFLOPs
933 - 1010 = -8%
Tex fillrate
48,160 - 50,640 = -5%
memory bandwidth
141.7 - 127 = +11.5%
No it doesn't, there's clearly examples of the 280 winning at 2560. Its also hilarious that you continue to consider benches with < 1 FPS difference as wins.Originally posted by: Azn
GTX 280 pulling away? Where? :laugh: GTX 275 beats GTX 280 in all of anand's benches.
TreiberversionenNo here knock yourself out. http://www.computerbase.de/art...nitt_performancerating
You do realize you're trying to argue both sides of the issue right? Either bandwidth is significant, or it isn't. I've already said it isn't significant on GT200 as I've clearly demonstrated a 27% decrease in bandwidth results in a 3% difference in performance. The fact a GTX 275 can (as the 260 before it) overcome any non-existent bandwidth limitation to come close or exceed a GTX 280, with much greater bandwidth, absolutely proves my point. Once again, you can't have it both ways although you've consistently argued both sides claiming it does or doesn't matter, depending on how wrong you are at any given point.You make me laugh. Did I say bandwidth was more important than SP or TMU? Of course not but it's only in your interest to say ignorant things to derail anyone who challenges your theories. I just said it's part of the GTX 280 equation since you said bandwidth didn't matter on GTX 280. Even in your own benches you've tested your GTX 280 you had minimum frame rates dropping as much 25%. Average frame rates dropping by 5%. To say bandwidth doesn't matter even on a bandwidth saturated card like GTX 280 is quite ignorant when you have tested the theory yourself.
Uh, no memory clocks have no impact on the ROP whatsoever, that's why they have independent clockspeeds. The clockspeed of the memory controllers themselves are tied to the core/ROP domain and are not impacted at all by memory frequency.Clocking it same on the core fine but what about bandwidth or the bandwidth saturation on the ROP itself. You would have to clock the GTX275 1264mhz on the memory to get same bandwidth as GTX 280. Even then High memory clocks has a effect on the ROP since they are tied down to the ROP itself so it would give advantage to GTX 275. This wouldn't be accurate but you would have to know that memory controllers are tied down to ROP to understand this. I've already mentioned this to you numerous times but you never could understand the concept of memory bandwidth effecting how ROP performs. Even then you don't even have the data to prove anything.
LMAO, no, I've never claimed bandwidth mattered, so trying to lump bandwidth in with ROP performance to make the numbers bigger in your favor isn't going to work. If anything the fact bandwidth doesn't hold the GTX 275 back compared to a part like the GTX 280 with more bandwidth further proves my point.Since you said ROP makes the biggest performance difference GTX 280 should beat GTX275 with combination of ROP and bandwidth but it doesn't. GTX 275 beats GTX 280 by 5% on average.
Which brings us back to my point. Which variable do you think impacts performance the most? I've maintained ROP and core domain, I guess you're claiming "the rest" lol.1. GTX280 | 2. GTX 275
ROP fillrate
19264 - 17724 = +8%
GFLOPs
933 - 1010 = -8%
Tex fillrate
48,160 - 50,640 = -5%
memory bandwidth
141.7 - 127 = +11.5%
You know if you blog about that 2900pro enough, someone might believe you about AMD parts being good overclockers historically. Looks like general concensus also seems to think the 4890's overclocking headroom is a pretty big deal for AMD parts.Originally posted by: SlowSpyder
I think it's quite common for 4870's to get an additional 10% overclock. Hardly an incredible overclock, but not like there's nothing left in the tank either. Seems to be about what the GTX285's get on average, 10%.
4850's overclock well over all. I've seen plenty of 3870's with 10%. My 2900 Pro overclocked 40%. It's not too hard to find good and bad examples from both camps.
Originally posted by: SlowSpyder
Originally posted by: error8
Originally posted by: SlowSpyder
My 2900 Pro overclocked 40%.
That doesn't count, since it was an underclocked XT all along. :laugh:
If the 4850 and GTX260 count (they are underclocked versions of other higher end cards) than I count the 2900.
Originally posted by: chizow
PCGH is actually one of the best reviews as it shows the 4870 and 4890 scale identically with clockspeed, clearly showing no enhancements or shader optimizations, clockspeed bumps for the increased performance by clocking a 4870 to 850MHz.
Originally posted by: error8
It really makes me wonder, why hasn't ATi released the 4890 with 950 mhz core clock right from the start, since it's pretty clear that all of the reviewed samples are topping 970-1000mhz? It would have struck right into the GTX 285, making this card more "logical" then it is .
Originally posted by: yacoub
For me the debate is whether I want to put up with ATi's typically flaky drivers for a couple % better performance when it's working right, when the GTX-275 is already a huge leap in performance over my 8800GT and would have the expectation of stable and consistent performance just as I've had with my 8800GT for the past 18 months and my 7900GT before that, compared to my X800XL and Radeon 9800 Pro and 9600 Pro that all had texture flashing or stability issues in some form or another.
Originally posted by: error8
Originally posted by: yacoub
For me the debate is whether I want to put up with ATi's typically flaky drivers for a couple % better performance when it's working right, when the GTX-275 is already a huge leap in performance over my 8800GT and would have the expectation of stable and consistent performance just as I've had with my 8800GT for the past 18 months and my 7900GT before that, compared to my X800XL and Radeon 9800 Pro and 9600 Pro that all had texture flashing or stability issues in some form or another.
Yeah, well, lot of time has passed since your X800XL and Radeon 9XXX days and flashing textures aren't available anymore. :laugh:
But still, I would probably choose Nvidia too, if I would have had serious problems with ATi in the past.
Originally posted by: yacoub
But yes, whatever works for you is the right answer. I'm thinking the 275 will be the right answer for me when I'm ready to buy. Hopefully it is. =)