R260/270/280/290/290x Review thread

[24601]

I'll just wait for a cheap 290 and put an Accelero cooler on it. I don't understand all the fuss people are making about the reference coolers. Who cares about reference coolers......they've aways been lousy. Also Nvidia reference coolers are a bit noisy. I want dead quiet coolers. And mostly that goes best with 120mm fans.

<<<<<<<<<<<<-========= Apparently I don't exist?

People care about blower performance. It matters in far more situations than one would like to believe.

 

[PCboy]

What a joke. All of these complaints about the noise/power consumption/random irrelevant nonsense to defend their overpriced purchases because they shot themselves in the foot for not waiting. The 100 dollars I'll save from buying a 290 could even get me a waterblock, let alone pay for my power bill that I would save if I were using more "efficient" videocards.

Hexen said it best, you guys are just Sunday gamers. Real gamers would be wearing headsets anyways. Get over yourselves.


Use of inflammatory language is not permitted.

-Rvenger

 

[Gikaseixas]

Anyone have an exact date of the >>real<< 290x and 290 launch?

AKA, when the things ship with an actual cooler instead of one powered on imagination?

This is the longest I've ever been strung along in a GPU launch, and that's not a good thing.

My 7970s and 7950s aren't going to replace themselves.

I also pray to the gods of video card manufacturing to have a bountiful harvest of AIB blower coolers. (Hah, as if that's going to happen)

Right after GTX780ti drops LOL

 

[24601]

What a joke. All of these complaints about the noise/power consumption/random irrelevant nonsense to defend their overpriced purchases because they shot themselves in the foot for not waiting. The 100 dollars I'll save from buying a 290 could even get me a waterblock, let alone pay for my power bill that I would save if I were using more "efficient" videocards.

Hexen said it best, you guys are just Sunday gamers. Real gamers would be wearing headsets anyways. Get over yourselves.

Real enthusiasts overclock to the maximum, and that would mean 100% fan speed and +50% power levels on these cards.

If you don't do that, you're not a Real enthusiast.

Play your 30 second time demos more.

Spending 350+ USD on water cooling means you didn't care about price performance to start with, so that's a non-starter.

 

[24601]

Right after GTX780ti drops LOL

I really hope so, since most people are saying a month or so.

 

[3DVagabond]

http://www.tomshardware.com/reviews/radeon-r9-290-review-benchmark,3659.html

Tomshardware bought 2 retail 290X's and both of them perform considerably worse than the golden samples that they were clearly sent by AMD. This raises some serious concerns. All of the reviews I saw of the original card put it ahead of the 780, yet if the retail cards are actually performing just a smidge over a 770 we are getting ripped off. It does appear AMD has been dishonest on purpose and I hope other reviewers take note of this and get retail cards to test. This could be a gigantic scandal.

With so much riding on the amount of throttling the card does its a big concern. The after market cooler they use however seems to be awesome, should solve the problem and make an excellent performing card to boot. But you need to factor that into the price of the card or watercool the darn thing.

Frame variance is still a bit of a concern looking at the pcper.com results for crossfire. Its not a sea of orange in the background but its still at levels where people can tell its there.

These cards are turning out to be a trade off, noise to performance and at crossfire variance to outright frame rate. Its not a clear cut win really, the rough comes with the awesome.

I can't find where they bought 2 retail cards. I see where they have one. Also, I don't see where they have said that it's apparent AMD is being dishonest. It's a 20 page article, maybe I'm just missing it?

I did find this

Update: As is Tom's Hardware policy, we shared these potentially problematic findings with AMD prior to publication, and the company insists something is wrong with the retail-purchased cards I tested. We will continue investigating and, if any additional news becomes available, update this story.

So there could be an issue with the particular card? Is that what they are actually saying?

 

[BrightCandle]

I can't find where they bought 2 retail cards. I see where they have one. Also, I don't see where they have said that it's apparent AMD is being dishonest. It's a 20 page article, maybe I'm just missing it?

In the conclusion they say the following (all bolding is Toms bolding):

However, the two retail Radeon R9 290X boards in our lab are both slower than the 290 tested today. They average lower clock rates over time, pushing frame rates down. Clearly there&#8217;s something wrong when the derivative card straight from AMD ends up on top of the just-purchased flagships. So who&#8217;s to say that retail 290s won&#8217;t follow suit, and when we start buying those cards, they prove to underperform GeForce GTX 780? We can only speculate at this point, though anecdotal evidence gleaned from our experience with R9 290X is suggestive.

Then further down they also say the following:

I have two issues with this. First, at 47% duty cycle, the fan is too loud. It&#8217;s obviously not as bad as the 290X&#8217;s Uber mode, but I don&#8217;t see any compelling reason to compromise acoustics when quieter solutions exist. AMD points out that you can turn the fan down if you want, and that's true, but you'd watch 290's performance erode at the same time. Second, I simply don&#8217;t trust the numbers I&#8217;m getting from the 290 we have on-hand to review. Even if it&#8217;s a total fluke that the R9 290X cards we have are so diametrically opposed, the mere existence of this much variance means Radeon R9 290 is either as fast as a GeForce GTX Titan and priced phenomenally or somewhere behind a retail R9 290X, just ahead of GeForce GTX 770, and priced to slot into the market (unspectacularly). I&#8217;m not comfortable making a recommendation one way or the other on 290 until we see some retail hardware.

And further down in the conclusion he is still concerned

Less consistency, tighter pricing, more noise...let's just say I'm more wary this time around.

I am the one saying it looks like AMD is being dishonest. Because all of the reviews I read based on review samples put the card ahead of the 780 most of the time, and yet two retail cards both perform worse. That smacks of cherry picked golden samples sent to reviewers. As a retail purchaser it looks like you can expect worse performance and more noise than these guys are seeing, and they are tell us these cards are too loud.

So you need to factor in the $100 aftermarket cooler into the price to bring back the consistency in the performance and to get what you paid for and what you are seeing in the reviews. Makes it much less attractive than it first appears on its performance numbers alone.

 

[24601]

I can't find where they bought 2 retail cards. I see where they have one. Also, I don't see where they have said that it's apparent AMD is being dishonest. It's a 20 page article, maybe I'm just missing it?

I did find this


So there could be an issue with the particular card? Is that what they are actually saying?

Page 2

The card that AMD sent to me is a stallion. Even if you get it nice and hot before running a test, bringing it down off of that 1000 MHz &#8220;wishful thinking&#8221; spec, it&#8217;s still faster than GeForce GTX 780, and oftentimes GeForce GTX Titan. But the Radeon R9 290X I bought from Newegg is a dud. It&#8217;ll drop to 727 MHz and stay there&#8230;and the reference cooler still can&#8217;t cool it fast enough. The result is that it violates its 40% fan speed ceiling as well. The craziness, then, is that my R9 290 press board is typically faster than my R9 290X retail card. In the benchmarks, you&#8217;re going to see numbers for all three.

The reason is obviously because the press got golden samples in the sense of low voltages. Take a look at the hilariously low voltage of the Anandtech samples and compare them to what people have actually been experiencing on their cards. The golden samples boost to a maximum of 1.11v while the forum goer's go to 1.3v. Absolutely laughable.

 

[OCGuy]

Nice to see AT and Toms keep credibility, while hardforum shows it is purely a AMD advertisement.

 

[Grooveriding]

Well you know AMD just released another decent card when you get the usual trolls doing their tryhard schticks.

I'm wondering if the GTX 780 will end up dropping further in price. It looks ridiculous all over again with this second AMD 290 series release. Another $100 cut off the price seems in order and I could go for a third once the price falls to $400 or less for GTX 780.

 

[VulgarDisplay]

Nice to see AT and Toms keep credibility, while hardforum shows it is purely a AMD advertisement.

It's nice to see that I have never witnessed such trumpeting of Nvidia retail cards in a review. I'm sure this is the next smoothness/FCAT FUD though.

Seriously, when has a review EVER used a retail card and trumpeted that fact in such a open manner? I can't remember another review like that. I'm positive all the Nvidia cards in reviews are perfect indication of stock duds that somehow snuck through validation.

 

[Gikaseixas]

So there could be an issue with the particular card?

Anand forum member bought one and pushed it to 1100mhz, says card stays there @ 75c while playing BF4
http://forums.anandtech.com/showthread.php?t=2351689

So it might be a dud Toms got. They should RMA it and try again before draw a conclusion

 

[3DVagabond]

In the conclusion they say the following (all bolding is Toms bolding):



Then further down they also say the following:



And further down in the conclusion he is still concerned



I am the one saying it looks like AMD is being dishonest. Because all of the reviews I read based on review samples put the card ahead of the 780 most of the time, and yet two retail cards both perform worse. That smacks of cherry picked golden samples sent to reviewers. As a retail purchaser it looks like you can expect worse performance and more noise than these guys are seeing, and they are tell us these cards are too loud.

So you need to factor in the $100 aftermarket cooler into the price to bring back the consistency in the performance and to get what you paid for and what you are seeing in the reviews. Makes it much less attractive than it first appears on its performance numbers alone.

Cheers. Thanks for finding the 2 cards for me.

Page 2



The reason is obviously because the press got golden samples in the sense of low voltages. Take a look at the hilariously low voltage of the Anandtech samples and compare them to what people have actually been experiencing on their cards. The golden samples boost to a maximum of 1.11v while the forum goer's go to 1.3v. Absolutely laughable.

I saw what you posted about the one Newegg card. I didn't see what Brightcandle was good enough to point out. I'm going to guess there are QC problems with the cooler mounting on the retail cards Tom's has. Not good if these cards made it out that way. It might explain the shortage of cards at retail, if that's the case. I imagine they are going to have to have a close look at teh cards in the retail supply chain. Not good!

 

[24601]

Anand forum member bought one and pushed it to 1100mhz, says card stays there @ 75c while playing BF4
http://forums.anandtech.com/showthread.php?t=2351689

So it might be a dud Toms got. They should RMA it and try again before draw a conclusion

Do you even V-sync bro?

Also I doubt that was at 40% fan speed with the reference cooler.

 

[selni]

Nice to see AT and Toms keep credibility, while hardforum shows it is purely a AMD advertisement.

Just like they advertised crossfire stuttering? It's probably just a difference in what people tolerate noise wise, and is it surprising to see [H] on the upper end of that scale?

 

[horza]

..

 

[selni]

Anand forum member bought one and pushed it to 1100mhz, says card stays there @ 75c while playing BF4
http://forums.anandtech.com/showthread.php?t=2351689

So it might be a dud Toms got. They should RMA it and try again before draw a conclusion

That link is running the fan at at least 65%. I'm pretty sure Tom's isn't doing that, for obvious reasons.

 

[ShintaiDK]

Wow, looking on Anandtechs 290 review. The 290 just made the 290X useless on reference designs. Its simply another giant statement of the cooling disaster. And the 290 is still pegged at the throttlemax temperature all the time. Had AMD just got the cooling right. Beaten by their own product this time with a 150$ penalty. Thats simply shooting yourself in the foot with an unseen huge caliber. Talk about lost oppotunity. Real shame.

 

[BrightCandle]

Who has one of the retail 290X's. Can you please let us know what voltage your card runs at. We really need to know if AMD has been sending golden samples to the review sites, if so we need to be very wary about this card until we see more reviewers notice and test retail cards.

 

[Silverforce11]

OCGuy, you are ridiculous bro, couldn't find a harder NV troll than you, to openly accuse [H] of being AMD biased.. (after all the [H] prior bashing of AMD for their CF issues repeatedly) disgusting behavior.



Personal attacks and use of inflammatory language are not permitted and against the forum rules.

-Rvenger

 

[el etro]

Nice to see AT and Toms keep credibility, while hardforum shows it is purely a AMD advertisement.

[H] doing PR to AMD?? Noooooo....

 

[wand3r3r]

Nice to see AT and Toms keep credibility, while hardforum shows it is purely a AMD advertisement.

Is this the first GPU launch you've followed? It certainly isn't theirs and every launch people seem to claim the reverse (depending on their preferred brands placement). It wasn't long ago people were claiming the reverse.

Additionally people were just claiming AT is AMD biased, now they aren't since they criticized AMD so they just got all "credibility" back? Let's switch the goalposts, it's funny to watch the scramble.

There's a lot of people pretty desperate to spread FUD. $400 with noise vs. $500 ($650 without the release of the cards) is a clear win to unbiased customers. Aftermarket cards are still the main thing I'm concerned with.

The funny thing is I certainly didn't see much uproar from seeing golden sampled 680's hitting 1300 MHz which in practice is almost unheard of.

 

[AtenRa]

If reference cooler is loud for you, you can always wait for the custom cards from ASUS, MSI, GigaByte and the rest. This card is the King of Performance/price in high-end territory period.

 

[BrightCandle]

The funny thing is I certainly didn't see much uproar from seeing golden sampled 680's hitting 1300 MHz which in practice is almost unheard of.

The difference here is that these are not overclocked results, they are default stock clock results that vary by nearly 40% in performance from card to card. The guaranteed performance level is a long way below the performance shown by reviewers and the noise level will also be lower than the usual cards (because the GPU ramps up its speed once it hits its minimum clocks). That is not the same thing as how much overclocking headroom a card has, or what its boost clocks are. With Nvidia's system there is a guarantee of a certain performance level which is within a few percentage points of the very best review cards. There isn't a 40% difference in clock speed.

I see a big difference with that. I take exception to the boost clock system as well, variance in 780's bothers me as well. the difference here however is striking, and based on all the reviews looks like AMD is being actively manipulative.

Lets put it another way. If you were a customer and you bought a 290X for $550 because it was faster than the 280X but obviously more expensive. You don't intend to overclock, you want it to last a while so you unwrap it and install it. Its not a golden chip, so it clocks down, runs louder and performs about the same as a 280X. That is wrong, no way to spin that as anything but wrong. If in the process it also turns out that AMD shipped specially selected rare cards that performed considerably better to reviewers to make it look like it dramatically outperformed the 280X and even beat Nvidia's Titan (a card that incidentally no retail purchaser could reasonably get) well then that would be grounds for a class action lawsuit. You see the problem yet or not?

 

[BrentJ]

[H] doing PR to AMD?? Noooooo....

Before last month we were called NVIDIA biased. I guess we are doing our job right.