[VC]AMD Radeon R9 390X WCE Speculation Thread

[96Firebird]

I am glad you included the you believe how it works part :thumbsup:

if that is how it was, with no reference models available for testing. why not forgo the reference tests? just do best of the best AIBs from both sides, voila. This was exactly what I wanted when I was shopping around for a gpu back in nov/2014. why used a custom AIB and down clock it? from the power usage charts, they weren't down clocked AIBs

Because there are still cards out there that run at reference clocks, even though that may not be the card the reviewer received for their launch review.

 

[Face2Face]

"Another shortcoming of the cooler, which looks fantastic with its black powder coat, by the way, is fan noise. While not terribly noisy, it definitely emits more noise than NVIDIA's other recent releases and roughly matches the Radeon R9 290X noise." Source

That's very odd. I wonder if NVIDIA has made changes regarding the RPM of their blower fan? I remember how quiet the reference cooler was on my GTX 780 at 75% fan speed. I wonder if it's maxed out @ 100% while gaming?

 

[monstercameron]

Maybe it was the moment it was launched before drivers were tuned properly, but the narrative for the last year from most publications is that the 290X is the better card for "high end performance".



It would be nice if we could fast-forward to 6 months from now and compare Titan X and 390X with refined drivers, but I am not a Time Lord.

Ot/ I hate the new doctor so much...

Slightly ot/ hopefully d3d12 drivers situation should be better and monthly drivers and hot fixes will be a thing of the past.

 

[Headfoot]

Who is using the G1?

And TechSpot is indeed using the WF3 290X, but at reference clocks.



Source

Good catch

 

[destrekor]

That's very odd. I wonder if NVIDIA has made changes regarding the RPM of their blower fan? I remember how quiet the reference cooler was on my GTX 780 at 75% fan speed. I wonder if it's maxed out @ 100% while gaming?

The article strangely words that, because their own chart doesn't really suggest that at all.

They list 40dBA for the Titan X under load. They then list the 290X (Quiet) at 40dBA, however the 290X (Uber) is at 50dBA.

The 295X2 is listed at 41dBA.

Now, non-reference 290X coolers perform much better than the standard Uber fan profile, I saw the 290X Lightning listed at 42dBA somewhere.

However, the quiet profile fan on the 290X is not a good comparison IMHO, and there is a world of difference between 40dBA and anywhere close to 50dBA.

 

[xthetenth]

"Another shortcoming of the cooler, which looks fantastic with its black powder coat, by the way, is fan noise. While not terribly noisy, it definitely emits more noise than NVIDIA's other recent releases and roughly matches the Radeon R9 290X noise." Source

Wait, when did roughly matching the R9 290X noise become not terribly noisy? If that's reference R9 290X, what is happening here? That's really bad review writing.

I watched the PCPer's live coverage of the Titan X with NVIDIA's Tom Petersen yesterday and he took a couple tiny digs at water cooling being a negative solution that "consumers don't want". He never mentioned anyone by name, just that it's something that they "would never consider" for a premium product like the Titan X.

Well, if that thing performs, this consumer's going to not want one right into his case. Cool quiet and good looking graphics is a much better product than a good looking blower. I guess they figure there's a crowd that want to be reminded they have a "premium" product every time the fan ramps up.

 

[Silverforce11]

"Under load, the GeForce GTX Titanium X comes to a volume of 47.5 decibels. This is significantly louder than the GeForce GTX 980 and a little louder than the GeForce GTX 780 Ti. In maximized settings the graphics card reached 50.5 decibels. and is disturbingly loud."

http://www.computerbase.de/2015-03/nvidia-geforce-gtx-titan-x-im-test/7/

Still less than R290/X in uber though, that's intolerably loud.

It's just a bit louder than 780ti which is fine. It only gets very loud once OC. Note that their "Max" isn't even a big OC with offsets, its just lifting temp/power target to max allowed while leaving offsets at default, IIRC.

 

[Headfoot]

"Under load, the GeForce GTX Titanium X comes to a volume of 47.5 decibels. This is significantly louder than the GeForce GTX 980 and a little louder than the GeForce GTX 780 Ti. In maximized settings the graphics card reached 50.5 decibels. and is disturbingly loud."

http://www.computerbase.de/2015-03/nvidia-geforce-gtx-titan-x-im-test/7/

Still less than R290/X in uber though, that's intolerably loud.

It's just a bit louder than 780ti which is fine. It only gets very loud once OC. Note that their "Max" isn't even a big OC with offsets, its just lifting temp/power target to max allowed while leaving offsets at default, IIRC.

Of course, with the 780 Ti and the 290/x you can choose to buy a model with a better cooler for as little as $0-10 more

 

[boozzer]

Because there are still cards out there that run at reference clocks, even though that may not be the card the reviewer received for their launch review.

then why not include both reference + whatever clocks the custom card came with in the charts? there is really no reason not to include both. then I can see the power usage. reviews should also include the info about downclocked AIBs in their testing methodology page just for clarification purposes. just 1 extra test. I, as a consumer, want as much info as possible. Don't you agree?

 

[waffleironhead]

then why not include both reference + whatever clocks the custom card came with in the charts? there is really no reason not to include both. then I can see the power usage. reviews should also include the info about downclocked AIBs in their testing methodology page just for clarification purposes. just 1 extra test. I, as a consumer, want as much info as possible. Don't you agree?

Maybe no one game AT a free card to review? That seems their excuse of why they dont review things these days. No free sample from the company=no review.

 

[redzo]

"Another shortcoming of the cooler, which looks fantastic with its black powder coat, by the way, is fan noise. While not terribly noisy, it definitely emits more noise than NVIDIA's other recent releases and roughly matches the Radeon R9 290X noise." Source
NV left the door wide open for AMD to take advantage of better price/performance and superior cooling and noise levels. Hopefully AMD executes.

Stock 290x cooling was so crappy, it achieved industry's lowest possible denominator award. It would take nvidia a long time to match that "performance". It's AMD's fault for setting the bar so low. Lesson learned, nvidia internal paper: "How Not to screw up a good GPU with a cooling fan".
The funny thing is: 290x stock cooling left AMD so brain damaged that they stuffed a water cooling solution on the 390x for nothing, when they could have just as easily gotten out without it. I guess we're gonna find out soon if this is the case.

 

[garagisti]

Stock 290x cooling was so crappy, it achieved industry's lowest possible denominator award. It would take nvidia a long time to match that "performance". It's AMD's fault for setting the bar so low. Lesson learned, nvidia internal paper: "How Not to screw up a good GPU with a cooling fan".
The funny thing is: 290x stock cooling left AMD so brain damaged that they stuffed a water cooling solution on the 390x for nothing, when they could have just as easily gotten out without it. I guess we're gonna find out soon if this is the case.

For reference, all you need to do is consider VaporX 8gb model of 290x. It is one the coolest, quietest card in the series, powered by a fan, and reported TDP is about the same. So i don't see no reason why it wouldn't be possible.

 

[96Firebird]

then why not include both reference + whatever clocks the custom card came with in the charts? there is really no reason not to include both. then I can see the power usage. reviews should also include the info about downclocked AIBs in their testing methodology page just for clarification purposes. just 1 extra test. I, as a consumer, want as much info as possible. Don't you agree?

That is what AT and TPU do... Are you talking about another review site?

For the 970 launch review:

Moving on, for the purposes of our testing we will be looking at both the GTX 970 FTW in its shipping configuration and in a reference clocked configuration. EVGA has given us the reference GTX 970 vBIOS to flash to this card (taking advantage of the triple BIOS feature), allowing us to turn it into a standard GTX 970 for that part of our testing.

Source

Charts

TPU doesn't tell how they get the results labeled "GTX 970 4GB", but the first page has the reference clockspeeds listed. I can only assume they clocked an aftermarket card down to reference clocks, maybe you can ask W1zzard if you're curious...

Charts

Same thing happened for the 285 launch reviews...

Our R9 285 sample meanwhile is Sapphire’s R9 285 Dual-X OC. As this is a factory overclocked model, for the purposes of our testing we will be testing this card at both its factory clockspeed (965MHz/5.6GHz) and the R9 285 reference clockspeed (918MHz/5.5GHz) by underclocking our card to the appropriate clockspeeds.

Source

 

[RussianSensation]

The funny thing is: 290x stock cooling left AMD so brain damaged that they stuffed a water cooling solution on the 390x for nothing, when they could have just as easily gotten out without it. I guess we're gonna find out soon if this is the case.

You are looking at it wrong. The reason to use water cooling is because it's simply a superior performance solution to any reference blower in the world - lower noise levels and temperatures and overclocking with reduced penalty for increased level of noise and temperature that otherwise occurs under air.

295X2 runs 10-20C cooler + quieter than a Titan X while using ~450-500W of power because of a single AIO CLC. Titan X at stock speeds operates at 80C+.

I think a lot of people don't understand just how much superior AIO CLC is to air. If you overclocked Hawaii to the max like that on a Titan X's blower, your temperatures would probably be 90C and crazy loud.

Max overclocked Hawaii uses a lot of power.

Despite that, a single 120mm AIO CLC keeps the card at 80C max.

That means a theoretical 300W 390X would run cooler and quieter than a 250W Titan X and easily cope with 350-400W overclocking. With Titan X's overclocking, noise levels skyrocket to uncomfortable levels. Just because the Titan X's reference blower is better than 290X's, doesn't mean it's a great solution at all. In fact, it isn't -- it's loud & can't maintain high boost for extended periods of time in overclocked states. That's not a knock against Titan X itself, but simply a fact that blowers by design are inferior in performance, noise levels and overclocking to the best after-market open air solutions or AIO CLCs.

 

[boozzer]

@firebird I am talking about reference + oc AIBs for gpus reviews use in their charts, not just the gpu that is being reviewed.

ex chart: 980 review

980 reference
980 AIB
970 reference
970 AIB
290x reference
290x AIB
290 reference
290 AIB

do the same for any other gpu they wish to include in the charts.

 

[Cookie Monster]

You are looking at it wrong. The reason to use water cooling is because it's simply a superior performance solution to any reference blower in the world - lower noise levels and temperatures and overclocking with reduced penalty for increased level of noise and temperature that otherwise occurs under air.

295X2 runs 10-20C cooler + quieter than a Titan X while using ~450-500W of power because of a single AIO CLC. Titan X at stock speeds operates at 80C+.

I think a lot of people don't understand just how much superior AIO CLC is to air. If you overclocked Hawaii to the max like that on a Titan X's blower, your temperatures would probably be 90C and crazy loud.

Max overclocked Hawaii uses a lot of power.

Despite that, a single 120mm AIO CLC keeps the card at 80C max.

That means a theoretical 300W 390X would run cooler and quieter than a 250W Titan X and easily cope with 350-400W overclocking. With Titan X's overclocking, noise levels skyrocket to uncomfortable levels. Just because the Titan X's reference blower is better than 290X's, doesn't mean it's a great solution at all. In fact, it isn't -- it's loud & can't maintain high boost for extended periods of time in overclocked states. That's not a knock against Titan X itself, but simply a fact that blowers by design are inferior in performance, noise levels and overclocking to the best after-market open air solutions or AIO CLCs.

I dont think anyone is questioning the performance of an AIO CLC. But time and time again, many people have already pointed out why its a bad idea to have one as a reference design especially from a business perspective unless its a niche product like the R295X where it actually makes sense.

Perhaps the GM200 is pushing the blower cooler to its limits.. It would have been wise for them to go with a Titan Z type open air cooler (e.g. 7900GTX - central fan) but perhaps theyll do that next time.

 

[RussianSensation]

I dont think anyone is questioning the performance of an AIO CLC. But time and time again, many people have already pointed out why its a bad idea to have one as a reference design especially from a business perspective unless its a niche product like the R295X where it actually makes sense.

I strongly disagree. It's an amazing idea because:

1) There is no blower in the world that exists that can keep a 250W TDP card cool and quiet. Blowers simply cannot compete with an open-air cooled solution.

Max overclocked MSI Lightning R9 290X stays at 70-71C. I bet this card is using 350W+ of power but it stays cool and quiet, something that can't be said for a stock Titan X, nevermind overclocked Titan X!

2) It's hard to imagine any reviewer will be able to criticize the noise levels or temperatures or an AIO CLC flagship card because they should excellent given how well R9 295X2 fared in reviews.

3) The alternative would be to use a blower like Titan X's and have negative PR associated with the card being way too loud OR invest $ into an open air cooled design such as Sapphire Tri-X. However, why invest money into the latter when AIBs will provide their own solutions?

That's why AIO CLC is a smart reference design since it means launch reviews will be awesome. In fact, every NV card that is after-market (i.e., 970) is painted in a more positive light in reviews compared to reference cards (remember GTX460 FTW!). That's usually because after-market cards run cooler and quieter and can maintain higher boost clocks for longer periods. AIO CLC accomplishes much of the same and almost any modern case with a 120mm opening should be able to fit it. In reality, anyone who is OK with spending $600-700 for a flagship card should either already have a modern case or be able to buy a new one like NZXT 440 for $100-130. What's the big deal? I don't get how gamers are so reluctant to upgrade their PSUs and cases but they spend big bucks for latest and greatest GPUs that lose $200-400 of value in 15-18 months.....illogical.

Perhaps the GM200 is pushing the blower cooler to its limits.. It would have been wise for them to go with a Titan Z type open air cooler (e.g. 7900GTX - central fan) but perhaps theyll do that next time.

That's the point - you think AIO CLC for a reference card is a bad idea but there is no other viable alternative that's "better." If you make an open air cooled 300W TDP card, reviewers will criticize it for dumping all that heat inside a case and for sure not recommend it for smaller and mid-size cases with poor airflow. This is not a problem for AIO CLC because it exhausts most of the heat out of the case. As a result, it actually makes AIO CLC more universal for smaller and mid-size cases. Given how fast CPU AIO CLCs took in the last 5 years, AMD is taking a great business risk to introduce AIO CLCs to GPUs which actually benefit from such cooling tech more since they use way more power than CPUs.

Also, the Titan Z's cooler isn't good at all. First of all, a 375W Titan Z throttles with that cooler.

Second of all, why would AMD create yet another open air cooled solution when companies like Sapphire, MSI, Asus will have good options? AMD would just be competing with them and likely losing. AIO CLC on the other hand is a completely unique solution that has never been offered in a reference flagship card before. Worst thing, AMD goes back to air cooling for R9 490X series. Best case, ALL future flagship 250-300W TDP cards from AMD (and hopefully NV) ship with AIO CLC reference. Blowers are not great for 250W cards. Even HardOCP reported that 980 SLI throttled without a custom fan curve and those cards use way less power than a Titan X. Right now Titan X OCing is basically a no go for anyone who wants a quiet computer. With AIO CLC on a 390X, this won't be as big of an issue.

 

[Headfoot]

In this thread, people argue that a solution that is objectively proven to result in lower temperatures, lower noise, with a higher power consumption GPU is somehow worse.

What's next guys, how water cooling isn't a "premium design" like the Titan cooler because it doesn't use aluminum?

If you don't want a higher performing solution, you're not an enthusiast. Watercooling has long been very effective but expensive. Now it's just very effective.

 

[Silverforce11]

http://www.inno3d.com/products_detail.php?refid=192

People cannot be seriously bashing *that*. Insanely low temps, very low noise and heat out the case. Also pretty damn sexy too with a good design. For the same price.

 

[nurturedhate]

http://www.inno3d.com/products_detail.php?refid=192

People cannot be seriously bashing *that*. Insanely low temps, very low noise and heat out the case. Also pretty damn sexy too with a good design. For the same price.

Of course not, the color is different.

 

[3DVagabond]

I dont think anyone is questioning the performance of an AIO CLC. But time and time again, many people have already pointed out why its a bad idea to have one as a reference design especially from a business perspective unless its a niche product like the R295X where it actually makes sense.

Perhaps the GM200 is pushing the blower cooler to its limits.. It would have been wise for them to go with a Titan Z type open air cooler (e.g. 7900GTX - central fan) but perhaps theyll do that next time.

Well, if the AIO on the 390X vs Titan-X works out like the 295x2 vs Titan-Z I think it'll be fine.

According to what we have come to know, speaking directly with some leaders of the houses that develop solutions with GPU from AMD, it seems that the house of Sunnyvale has long been ready for the new cards R9 390X, but is waiting for the video drivers are practically perfect (a big announcement will be made ​​at Computex, as we predicted ).

While qusto means give away a few months ahead of sales to nVidia GTX 970 and GTX with 980, on the other hand will allow AMD to present a product almost entirely free of defects, so as not to commit the same mistake of which are been victims 7xxx series cards (Tahiti in the first place), so mistreated by users as journalists for drivers unripe

Contrary to what it has always done, now AMD does not point only to offer before the competitor product theoretically more powerful, but also more ealmente exploitable. The same strategy was adopted in the market for professional video cards for a few months, and look at the market shares gnawed at nVidia (not forgetting the closed contract with Apple for the Mac Pro) that approach is succeeding.

One last thing to note concerns the recent rumor appeared on the web, regarding the performance offered by the 390X. It seems that the performance of the 390X are much higher than those presented so far, and that is what we have seen so far, if not false, far from reality. Fingers crossed.

Just found this on an Italian site. Claims that 390X will be even stronger than the leaks are suggesting.

If this is true then it's as I said, they don't want to repeat the mistakes that Rory Read made. No half baked products being picked apart because of what the suits decided to do detracting from what the engineers made.

 

[nemesismk2]

can't wait to read some reviews of the AMD Radeon R9 390X.

 

[csbin]

Confirmed: R9 390X ready, but AMD expects that the drivers are mature

http://www.bitsandchips.it/9-hardwa...onte-ma-amd-aspetta-che-i-driver-siano-maturi

 

[therealnickdanger]

Confirmed: R9 390X ready, but AMD expects that the drivers are mature

http://www.bitsandchips.it/9-hardwa...onte-ma-amd-aspetta-che-i-driver-siano-maturi

Translation by Google:

One last thing to note concerns the recent rumor appeared on the web, regarding the performance offered by the 390X. It seems that the performance of the 390X are much higher than those presented so far, and that is what we have seen so far, if not false, far from reality. Fingers crossed.

I'll believe that statement when reviews start coming out. So the card is ready, but they are getting all the driver tweaks and profiles ready in advance? If true, that's very good news. AMD can't misfire on this one.

 

[Subyman]

If its just driver tweaking, then maybe launch will be sooner than I was thinking. No way its an April launch?