R9 290 *Complete* review list

Page 8 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

SlowSpyder

Lifer
Jan 12, 2005
17,305
1,001
126
Yea, we have a new metric. [Performance*smoothness/equalized_noise*Watt]*%of_industral_design


Didn't the reviewers show that the 290X is now 'smoother' in delivering frames than Nvidia's cards? Notice how suddenly those who were so loud about that before are no longer bringing it up?
 

MrK6

Diamond Member
Aug 9, 2004
4,458
4
81
Yea, we have a new metric. [Performance*smoothness/equalized_noise*Watt]*%of_industral_design
That's golden. If I had room in my sig it would go there.

The constant garbage the nvidia fanclub puts out in this subforum is nothing short of astounding.




Your last sentence was unnecessary.

-Rvenger
 
Last edited by a moderator:

Vesku

Diamond Member
Aug 25, 2005
3,743
28
86
Didn't the reviewers show that the 290X is now 'smoother' in delivering frames than Nvidia's cards? Notice how suddenly those who were so loud about that before are no longer bringing it up?

Yes, there is a trend. I even pointed it out during the "smoothness" blip. An interesting. but questionable on whether it is user impacting, difference between the brands shows up on a website then it gets immediately blown up by concerted posting on enthusiast forums. I'm sure AMD does it too, I'd be shocked if their marketing didn't latch onto the "Fermi Oven" opportunity. But imo the Nvidia propaganda is much more plentiful and hyperbolic, which makes it more annoying to me.

It wouldn't even bug me that much if the forum identities would be consistent and do such things as give occasional kudos to a company that actually improves what they've been criticizing. Would at least make PC tech forums a tad more civil.
 
Last edited:

monkeydelmagico

Diamond Member
Nov 16, 2011
3,961
145
106
AMD screwed up releasing so late. Few people will buy the reference card. Most will wait for the aftermarket cards. Will they be ready for Xmas? Not looking good.....
 

mingsoup

Golden Member
May 17, 2006
1,295
2
81
I'm really impressed from what I've read so far. I expected the 290 not to be noteworthy. or for the cooling to not help at all and you are left with a leafblower.

I'm also kinda anxious about gsync. However for the prices that AMD is pushing compared to nvidia for this level of performance, with the possibilities of mantle, its position in nextgen consoles, and! its advances in XFIRE [need to look into that more], I think an aftermarket 490 is in my immediate future. I suppose gsync could always be licensed. I'm still wondering if gsync will make GPU upgrades much less needed when 30fps feels like 60fps, effectively allowing you to upgrade half as much.

I listened to one of the aftermarket coolers they put on and its nothing compared to the reference even at full load. I couldn't even hear it. Whereas the reference was screaming.

I think this is the most exciting GPU launch I've seen in years.
 
Last edited:

f1sherman

Platinum Member
Apr 5, 2011
2,243
1
0
I'm really impressed from what I've read so far. I expected the 290 not to be noteworthy. or for the cooling to not help at all and you are left with a leafblower.

I'm also kinda anxious about gsync. However for the prices that AMD is pushing compared to nvidia for this level of performance, with the possibilities of mantle, its position in nextgen consoles, and! its advances in XFIRE [need to look into that more], I think an aftermarket 490 is in my immediate future. I suppose gsync could always be licensed. I'm still wondering if gsync will make GPU upgrades much less needed when 30fps feels like 60fps, effectively allowing you to upgrade half as much.

I listened to one of the aftermarket coolers they put on and its nothing compared to the reference even at full load. I couldn't even hear it. Whereas the reference was screaming.

I think this is the most exciting GPU launch I've seen in years.


How about the most configurable GPU?
http://www.pcper.com/reviews/Graphics-Cards/AMD-Radeon-R9-290X-Hawaii-Configurable-GPU

 

GaiaHunter

Diamond Member
Jul 13, 2008
3,634
180
106
Any performance numbers to go with those most dynamic of clock speeds vs the not so dynamic?

Just read the original pcper review of the R9 290X.



So basically 290X at an average of 836 MHz match a 780 at >966 MHz in Crysis 3 at 1080p.



Also of note is that users using uber mode (55% fan, the pcper is using 50% & 60%) won't see much if any throttling.

The conclusion as usual is:

If noise is a concern wait for AIB custom cards or do it yourself - that will reduce the noise, the temperature, maybe even some power reduction. Or just let you OC quite a bit.
 
Last edited:

blastingcap

Diamond Member
Sep 16, 2010
6,654
5
76
Just read the original pcper review of the R9 290X.

The conclusion as usual is:

If noise is a concern wait for AIB custom cards or do it yourself - that will reduce the noise, the temperature, maybe even some power reduction. Or just let you OC quite a bit.

My conclusion is that R9 290 series is pre-overclocked and possibly pre-overvolted as well. They are really pushing the chip hard right out of the box, just like NV pushed their GK104 chip in GTX 770. 7970 and 680 were not pushed nearly as hard so they had more headroom for post-purchase overclocking/volting.

PCPer ran the numbers hot vs cold and surprise surprise you need 50-60% fan speed to prevent the hot run from sinking in performance. http://www.pcper.com/reviews/Graphi...figurable-GPU/Cold-versus-Hot-R9-290X-Results

"The Run 2 graph shows us that 20% and 30% levels are just simply not going to work on the 290X. Take a look at the green line that represents 20% - it is actually running OVER 20% the entire time, and in fact runs over 30% and hits 44% towards the end! The same is true for the 30% setting. Clearly the R9 290X is not able to maintain its base clock of 727 MHz with fan speeds under 40% and thus we now know why AMD selected that option for its "quiet" profile. Rather than lower the clocks any more than 727 MHz, AMD instead overrides the maximum fan speed setting of the driver to lower temperatures and maintain a "base" performance level."

And "hot" was just 5-7 minutes of play. For longer periods of time I suspect you would need to ride 60%+ continuously.

You can argue all you want that buying an aftermarket card or Arctic Accelero or water will 'fix' your problem but that means:

- extra cost for Accelero, water, and possibly some of the better non-ref cards
- extra hassle for Accelero, water installation as well as possible warranty voiding issues
- problems with SLI/Xfire if open air coolers are packed close together
- still sitting at 95C, that can't be good for longevity no matter what AMD claims.. if not for your card then think of the waste heat being spewed into your case affecting your other components (with open-air coolers)
- you still have less headroom than, say, 7970 or 680 gave you. With those cards you got decent headroom even with reference cards and good cooling solutions, and better cooling gave you even more headroom. With R9 290/290X you get something that has been pre-oc/ov'd so it has almost no headroom left.. you need to crank to 60% fan just to get it to hit top stock speeds. So a card with better cooling merely gives you back a little headroom that you should have gotten in the first place.

I guess people who don't like messing with manual overclocks/volts might be ok with R9 290-series cards. Especially if they game with headphones. But for those of us who liked doing it ourselves, these GPUs are basically pre-oc/ov'd leaving virtually no headroom with reference cooler. Maybe you get back a few percent with good air coolers, but that's a far cry from cards like GTX 460 and HD 7950 that could oc 25%+ even with modest air coolers.
 
Last edited:

GaiaHunter

Diamond Member
Jul 13, 2008
3,634
180
106
PCPer ran the numbers hot vs cold and surprise surprise you need 50-60% fan speed to prevent the hot run from sinking in performance.
I guess people who don't like messing with manual overclocks/volts might be ok with R9 290-series cards. Especially if they game with headphones. But for those of us who liked doing it ourselves, these GPUs are basically pre-oc/ov'd leaving virtually no headroom with reference cooler. Maybe you get back a few percent with good air coolers, but that's a far cry from cards like GTX 460 and HD 7950 that could oc 25%+ even with modest air coolers.

I don't see the surprise.

The original reviews already talked about this.

And look at tomshardware or computerbase.de.

The drop in temperatures are significant.

There was one site (forgot which one) that strapped the HIS ICEQ X2 from a 280X and saw significant drops.

1150-1200 should normal with a decent air cooler. That is 15-20%, and we are already talking of performance similar to the 780.
 

blastingcap

Diamond Member
Sep 16, 2010
6,654
5
76
I don't see the surprise.

The original reviews already talked about this.

And look at tomshardware or computerbase.de.

The drop in temperatures are significant.

There was one site (forgot which one) that strapped the HIS ICEQ X2 from a 280X and saw significant drops.

1150-1200 should normal with a decent air cooler. That is 15-20%, and we are already talking of performance similar to the 780.

Are those hot runs or cold runs? Because I game for more than 5-7 minutes as a time.
 

Erenhardt

Diamond Member
Dec 1, 2012
3,251
105
101
Are those hot runs or cold runs? Because I game for more than 5-7 minutes as a time.

Do you really want to know, or are you just trolling? Because if you are interested, go look it up. Otherwise, just stop.



Enough of the personal attacks please.


-Rvenger
 
Last edited by a moderator:

blastingcap

Diamond Member
Sep 16, 2010
6,654
5
76
Do you really want to know, or are you just trolling? Because if you are interested, go look it up. Otherwise, just stop.

Are you always this rude? He did not post links and I don't recall reading anything saying 1200 was sustainable, maybe short term benchmarks but that is not like gaming for an hour.
 

SlowSpyder

Lifer
Jan 12, 2005
17,305
1,001
126
I don't know that a GPU that runs at ~800MHz is really pre-overclocked. Aren't people already running these at 1+GHz? I think we'll see a lot more headroom on these than some think. Just keeping the GPU from throttling will give you 25% more clock speed than the ~800MHz or so they seem to settle at! And only 5Gbps memory on the reference card? I think we'll see some very fast non-reference boards with ~1050MHz+ clocks and 6Gbps or higher memory. From what I've seen simply keeping these GPU's cool gives you massive boosts on the reference boards, and that's not even touching the factory clock settings.
 

GaiaHunter

Diamond Member
Jul 13, 2008
3,634
180
106
Are those hot runs or cold runs? Because I game for more than 5-7 minutes as a time.

Hot runs.

Some of the reviewers clearly mention that they warm up the cards.

BrentJ for [H] also confirmed that in his play the cards at uber stick at 1000 MHz, maybe with the exception of FC3.
 

GaiaHunter

Diamond Member
Jul 13, 2008
3,634
180
106
I don't know that a GPU that runs at ~800MHz is really pre-overclocked. Aren't people already running these at 1+GHz? I think we'll see a lot more headroom on these than some think. Just keeping the GPU from throttling will give you 25% more clock speed than the ~800MHz or so they seem to settle at! And only 5Gbps memory on the reference card? I think we'll see some very fast non-reference boards with ~1050MHz+ clocks and 6Gbps or higher memory. From what I've seen simply keeping these GPU's cool gives you massive boosts on the reference boards, and that's not even touching the factory clock settings.

Powercolor already has a 1050 card out, although it is an OC reference.
 

blastingcap

Diamond Member
Sep 16, 2010
6,654
5
76
I don't know that a GPU that runs at ~800MHz is really pre-overclocked. Aren't people already running these at 1+GHz? I think we'll see a lot more headroom on these than some think. Just keeping the GPU from throttling will give you 25% more clock speed than the ~800MHz or so they seem to settle at! And only 5Gbps memory on the reference card? I think we'll see some very fast non-reference boards with ~1050MHz+ clocks and 6Gbps or higher memory. From what I've seen simply keeping these GPU's cool gives you massive boosts on the reference boards, and that's not even touching the factory clock settings.

Big chips tend to run slower even if using same process as smaller chips, IIRC. Where is IDC when you need someone to explain chip physics?

One thing I do appreciate is that we are (FINALLY) back up to 512-bit memory bandwidth, something that hasn't happened in ages. Since, what, the GTX 285 I think? Something like that? And ROPs finally went up too, though current-gen games aren't exactly geometry bound most of the time. But maybe AMD knows something about next-gen console ports that we don't, so they preemptively beefed up memory and ROPs.

Hot runs.

Some of the reviewers clearly mention that they warm up the cards.

BrentJ for [H] also confirmed that in his play the cards at uber stick at 1000 MHz, maybe with the exception of FC3.

Ok, time will tell I guess. I'm still disappointed in oc headroom and have doubts that 1200 would be stable on anything other than water. I've seen so many people claim this or that for benching and stability but then it turns out less than stable if I try it on my cards. I can't just keep getting dog samples either given how many cards I go through, so I think my standards for stability may be higher than for most.
 

ocre

Golden Member
Dec 26, 2008
1,594
7
81
That's golden. If I had room in my sig it would go there.

The constant garbage the nvidia fanclub puts out in this subforum is nothing short of astounding.

Didn't the reviewers show that the 290X is now 'smoother' in delivering frames than Nvidia's cards? Notice how suddenly those who were so loud about that before are no longer bringing it up?

Yes, there is a trend. I even pointed it out during the "smoothness" blip. An interesting. but questionable on whether it is user impacting, difference between the brands shows up on a website then it gets immediately blown up by concerted posting on enthusiast forums. I'm sure AMD does it too, I'd be shocked if their marketing didn't latch onto the "Fermi Oven" opportunity. But imo the Nvidia propaganda is much more plentiful and hyperbolic, which makes it more annoying to me.

It wouldn't even bug me that much if the forum identities would be consistent and do such things as give occasional kudos to a company that actually improves what they've been criticizing. Would at least make PC tech forums a tad more civil.

I think AMD need commended for their improvements here. But by the tone of these statements, it almost sounds like an attempt at starting drama more than a thumbs up to AMD. I think AMD deserves praise for their improvement, without all that other mess.

I do believe that some my be forgetting where all this came from. It didnt start with the graphs but reviewers. Reviewers who where noticing hitching and inconsistent pacing. The frame charts came after to try to represent the hitching and compare the data in a meaning full way. There was a huge backlash from very vocal groups that said if you couldnt perceive the pacing issues then what is the problem. Well here we are full turn and it would be quite funny if those same people were now starting a fuss over frame times that are for the most part undetectable. I am not saying that any of those quoted are now making a fuss over something they once disregarded, just that it would be funny.

See, it can go both ways. But behind all the games and bickering there was a real perceivable frame pacing issues. There was pacing issues and studder that was noted. It took complicated measures to attempt to chart these variances in a meaningful way. In the beginning there was huge differences and massive issues in pacing frames. AMD has really done an extremely well job at correcting these issues, i would be the first to commend them for their work. As far as the charts go, AMD frame times look as good and even better than Nvidia's. But now we are talking about differences so small that it takes charts to see a difference. It is no longer studder that is noticed and a chart trying to represent that studder. We are past all that. Many people said if it is not perceivable then whats the big deal, the difference between the camps is so small now there is no big deal. Unless, all of a sudden these people want to try to make one out of it.

I, for one am glad to see the improvement. I am also glad to see that attention is still on smoothness to keep everyone in check. I think once there is a case where frame times gets choppy again, people will notice and it will be corrected. No matter which company or what card it is. This is good for everyone, no?
 

GaiaHunter

Diamond Member
Jul 13, 2008
3,634
180
106
Ok, time will tell I guess. I'm still disappointed in oc headroom and have doubts that 1200 would be stable on anything other than water. I've seen so many people claim this or that for benching and stability but then it turns out less than stable if I try it on my cards. I can't just keep getting dog samples either given how many cards I go through, so I think my standards for stability may be higher than for most.

I used to be like that has well, especially with CPUs, but recently I kept getting hardware that at least match the average OCs.
 

blastingcap

Diamond Member
Sep 16, 2010
6,654
5
76
I think AMD need commended for their improvements here. But by the tone of these statements, it almost sounds like an attempt at starting drama more than a thumbs up to AMD. I think AMD deserves praise for their improvement, without all that other mess.

I do believe that some my be forgetting where all this came from. It didnt start with the graphs but reviewers. Reviewers who where noticing hitching and inconsistent pacing. The frame charts came after to try to represent the hitching and compare the data in a meaning full way. There was a huge backlash from very vocal groups that said if you couldnt perceive the pacing issues then what is the problem. Well here we are full turn and it would be quite funny if those same people were now starting a fuss over frame times that are for the most part undetectable. I am not saying that any of those quoted are now making a fuss over something they once disregarded, just that it would be funny.

See, it can go both ways. But behind all the games and bickering there was a real perceivable frame pacing issues. There was pacing issues and studder that was noted. It took complicated measures to attempt to chart these variances in a meaningful way. In the beginning there was huge differences and massive issues in pacing frames. AMD has really done an extremely well job at correcting these issues, i would be the first to commend them for their work. As far as the charts go, AMD frame times look as good and even better than Nvidia's. But now we are talking about differences so small that it takes charts to see a difference. It is no longer studder that is noticed and a chart trying to represent that studder. We are past all that. Many people said if it is not perceivable then whats the big deal, the difference between the camps is so small now there is no big deal. Unless, all of a sudden these people want to try to make one out of it.

I, for one am glad to see the improvement. I am also glad to see that attention is still on smoothness to keep everyone in check. I think once there is a case where frame times gets choppy again, people will notice and it will be corrected. No matter which company or what card it is. This is good for everyone, no?

Apparently the way to fix stutter for good is G-Sync or something like it because you get to have V-Sync without the drawbacks of V-Sync. Your monitor refresh rate is chained to your video card's refresh rate. http://www.anandtech.com/show/7436/nvidias-gsync-attempting-to-revolutionize-gaming-via-smoothness
 

SlowSpyder

Lifer
Jan 12, 2005
17,305
1,001
126
Big chips tend to run slower even if using same process as smaller chips, IIRC. Where is IDC when you need someone to explain chip physics?

One thing I do appreciate is that we are (FINALLY) back up to 512-bit memory bandwidth, something that hasn't happened in ages. Since, what, the GTX 285 I think? Something like that? And ROPs finally went up too, though current-gen games aren't exactly geometry bound most of the time. But maybe AMD knows something about next-gen console ports that we don't, so they preemptively beefed up memory and ROPs.


I think they may have taken 32 ROP's as far as they could. I remember with the 7970 the AT review stated that AMD told them something along the lines of they made the ROP's much more efficient than the previous generations, so they stuck with 32. But I think you can only take that so far. Also, AMD likes to talk up 4k gaming with Hawaii, I imagine you need more than 32 ROP's to output all those pixels.

Yea, all else being equal I can see how a large chip would have to run at a slower speed at a given power level. But the ~800MHz they seem to settle at seems pretty low when you consider that the reference cards have a boost speed of 1GHz and Power Color even felt they could move that target up another 5% to 1050MHz. It may be that 800MHz sounds like such a low number to me when compared to other GPU's... I dunno.

When you consider that an 800MHz 290/290X is already as fast to faster than a 780GTX and they are set to boost to 1GHz from the factory, I have to think we'll see some very fast non-reference cards coming soon. 1.1GHz is 37.5% more clock speed than 800MHz. Add in plenty of room for faster memory and I think we have some amazing performance on the horizon.

I just hope the non-reference boards aren't too much over the $400 reference price tag. I didn't plan on really upgrading my 7970 as it is fine for the games I play at 19x12, but for what I can sell it for and what $400 buys today, I might consider it for the right card.

And yes, I agree we could always use more IDC posts. I see he has moved out of country (I assume for work), so I'm sure he's a busy guy these days.
 
Last edited:

GaiaHunter

Diamond Member
Jul 13, 2008
3,634
180
106
The GK110 is a bigger chip than Hawaii though.

Hawaii has a higher density of xtors - that also increases speed but also power consumption.
 

blastingcap

Diamond Member
Sep 16, 2010
6,654
5
76
Maybe it's a yield issue then, I'm not sure. I guess we'll know more as we get more data.
 

SlowSpyder

Lifer
Jan 12, 2005
17,305
1,001
126
The GK110 is a bigger chip than Hawaii though.

Hawaii has a higher density of xtors - that also increases speed but also power consumption.


I was going to post that, but I'm not sure if Hawaii is built on the same variation of TSMC's 28nm process. I thought there were a few variations now...? Maybe someone more knowledgeable can help me out here.
 
sale-70-410-exam    | Exam-200-125-pdf    | we-sale-70-410-exam    | hot-sale-70-410-exam    | Latest-exam-700-603-Dumps    | Dumps-98-363-exams-date    | Certs-200-125-date    | Dumps-300-075-exams-date    | hot-sale-book-C8010-726-book    | Hot-Sale-200-310-Exam    | Exam-Description-200-310-dumps?    | hot-sale-book-200-125-book    | Latest-Updated-300-209-Exam    | Dumps-210-260-exams-date    | Download-200-125-Exam-PDF    | Exam-Description-300-101-dumps    | Certs-300-101-date    | Hot-Sale-300-075-Exam    | Latest-exam-200-125-Dumps    | Exam-Description-200-125-dumps    | Latest-Updated-300-075-Exam    | hot-sale-book-210-260-book    | Dumps-200-901-exams-date    | Certs-200-901-date    | Latest-exam-1Z0-062-Dumps    | Hot-Sale-1Z0-062-Exam    | Certs-CSSLP-date    | 100%-Pass-70-383-Exams    | Latest-JN0-360-real-exam-questions    | 100%-Pass-4A0-100-Real-Exam-Questions    | Dumps-300-135-exams-date    | Passed-200-105-Tech-Exams    | Latest-Updated-200-310-Exam    | Download-300-070-Exam-PDF    | Hot-Sale-JN0-360-Exam    | 100%-Pass-JN0-360-Exams    | 100%-Pass-JN0-360-Real-Exam-Questions    | Dumps-JN0-360-exams-date    | Exam-Description-1Z0-876-dumps    | Latest-exam-1Z0-876-Dumps    | Dumps-HPE0-Y53-exams-date    | 2017-Latest-HPE0-Y53-Exam    | 100%-Pass-HPE0-Y53-Real-Exam-Questions    | Pass-4A0-100-Exam    | Latest-4A0-100-Questions    | Dumps-98-365-exams-date    | 2017-Latest-98-365-Exam    | 100%-Pass-VCS-254-Exams    | 2017-Latest-VCS-273-Exam    | Dumps-200-355-exams-date    | 2017-Latest-300-320-Exam    | Pass-300-101-Exam    | 100%-Pass-300-115-Exams    |
http://www.portvapes.co.uk/    | http://www.portvapes.co.uk/    |