[Sweclockers] Radeon 380X coming late spring, almost 50% improvement over 290X

[96Firebird]

Unfortunately, I'm almost certain that image is a photoshop of the 295X2 card.

https://www.tineye.com/search/502c7a28d29fc4705e3c7c40a6508480a6141c68/

Click compare under the second or third result and switch between the two. You can also see where the black doesn't line up on the bottom. And the cutoff for the PCI-E connectors is exactly where the 295X2's water lines run.

 

[KaRLiToS]

You must be about 99.9999% right FireBird (or 100%)

 

[96Firebird]

The shroud could look like that though, if the image in the OP's link is the actual shroud being used. Looks like that image first circulated the internet back in September, but has made a comeback as we get closer (hopefully) to launch.

 

[antihelten]

Unfortunately, I'm almost certain that image is a photoshop of the 295X2 card.

https://www.tineye.com/search/502c7a28d29fc4705e3c7c40a6508480a6141c68/

Click compare under the second or third result and switch between the two. You can also see where the black doesn't line up on the bottom. And the cutoff for the PCI-E connectors is exactly where the 295X2's water lines run.

For what it's worth the image originates from videocardz.com (http://cdn.videocardz.com/1/2014/11/Radeon-R9-390-concept-475x356.jpg), not that that really makes it anymore reliable (I agree that it looks like a photoshop)

 

[3DVagabond]

The 380X is 300W. So the 390X will be a 500W single GPU that'll burn out in 6 months then? Either way, it looks like I'll have to pass. :/ Absolutely pathetic.

Where are you getting 500W and that it'll burn out in 6 months?

 

[3DVagabond]

http://wccftech.com/r9-380x-coming-q2-2015/

Well, that pic is definitely fake. Unless we are to believe they can cool a 300w card with a single standard fan.

 

[antihelten]

Well, that pic is definitely fake. Unless we are to believe they can cool a 300w card with a single standard fan.

The two holes on the side are for the tubes of the AIO cooler

 

[Silverforce11]

It will be a variation of that since Asetek already released a public statement confirming as such a few months ago. It's gonna be a hybrid water cool part. 300W will be nothing to it since a single thin 120mm rad on the R295X2 is capable of keeping 2x R290X GPUs cool and quiet.

 

[SlickR12345]

Clearly a fake post. That website even states that all the information is just rumors from other sources, nice job OP creating a fake post with fake info.

I personally know that the next AMD graphic cards are going to be 20nm. So the fact that this mentioned 28nm set off the alarm bell.

 

[Blue_Max]

Well, if it WERE true... I like speed as much as anybody, but three hundred freakin' watts?? That's shamefully inefficient power use!

Sure hope it's baloney anyway.

 

[Silverforce11]

Do you realize efficiency is a measure of performance per watt for GPUs?

Would it be inefficient if it was 500W and 3x faster than R290X?

-_-

 

[RussianSensation]

I personally know that the next AMD graphic cards are going to be 20nm. So the fact that this mentioned 28nm set off the alarm bell.

Insider info from someone? How do you know for sure they will be 20nm?

 

[3DVagabond]

The two holes on the side are for the tubes of the AIO cooler

So where's the cooler?

 

[Mondozei]

By the way, if we assume a relatively optimistic 70% improvement over 290X that is still not amazing considering that by the time of the launch(likely around Computex, so early June), the 300-series would be just a few months short of two years after Hawaii was released.

So if we use a common rule of thumb of about 30% increase performance per year, that gives us 1.0*1.3*1.3=1.69, so almost 70%. That is entirely in line with expectations. Maxwell provided a 28% improvement, when you compare 780 to 980. I'm using Sweclocker's performance index, where they average a bunch of games and normalize performance. And that is really a somewhat unfair comparison since the 780 was Big Kepler while the 980 is a midrange GM204 arch.

If we get 20 nm on top of this, as well, then I'm not even sure if the 70% would be that revolutionary, then I'd say we'd need more than 100% improvement for it to really hit it out of the park, especially considering that you have HBM, too.

 

[RussianSensation]

By the way, if we assume a relatively optimistic 70% improvement over 290X that is still not amazing considering that by the time of the launch(likely around Computex, so early June), the 300-series would be just a few months short of two years after Hawaii was released.

So if we use a common rule of thumb of about 30% increase performance per year, that gives us 1.0*1.3*1.3=1.69, so almost 70%. That is entirely in line with expectations. Maxwell provided a 28% improvement, when you compare 780 to 980. I'm using Sweclocker's performance index, where they average a bunch of games and normalize performance. And that is really a somewhat unfair comparison since the 780 was Big Kepler while the 980 is a midrange GM204 arch.

If we get 20 nm on top of this, as well, then I'm not even sure if the 70% would be that revolutionary, then I'd say we'd need more than 100% improvement for it to really hit it out of the park, especially considering that you have HBM, too.

From a historical technology point of view, no it wouldn't be "revolutionary" since 60-70% more performance over 290X would be 2.2-2.3X faster than an HD7970Ghz but 3 years later, a similar feat R9 290X accomplished over the HD6970.

Similarly, 7970Ghz was 75-77% faster than an HD6970 1.5 years later (same link).

You are right that on an annualized basis tracking back from HD5870/480, GPUs increase about 30-35% per year and 75-100% performance increase every 2 years or so is now unobtainable. So, no it won't be revolutionary like X800XT --> X1800XT or 5900U to 6800U.

However, I think you gotta look at the timing vs. performance increase now because we are stuck on 28nm:

R9 290X / 780Ti = Nov 2013 = 100%
GTX980 = Sept 2014 (10 months) = 115-120%
R9 390X = June - July 2014 (9-10 months) = 160-170%

Even if R9 390X is 45% faster than an R9 290X, it will already be more impressive than a 980 given the context of time.

In your 780 to 980 example, 1.5 years passed and 980 provided just 32% at 1440p, and 34% at 4K. If 390X drops with 45% over the 290X by Summer of 2015, it will have already done A LOT better than 980 did vs. 780/780Ti. It's from this point of view that R9 390X delivering 60-70% would be absolutely mind-blowing after the lackluster gains 980 brought at $550-600 price levels. We would be back to similar gains that 7970Ghz delivered over the 6970 or 290X did over the 6970 given the context of time

I think some of us just can't believe the excitement that ensued with 970/980 cards because like you we expected 40-50% faster than 290X from a next gen flagship card. That's why to me a 980 is just a mid-range card that NV is milking now because they can and consumers keep falling for the same tactics that NV used with 680-> 780Ti transition. Other than power usage and HDMI 2.0, I am no longer impressed with the 970/980 cards as I was at launch as I got caught into the hype train and almost got dual Gigabyte 970s myself.

Also, 50-60% faster than a 290X puts us as 295X2 level of performance. That's very impressive to me from a single chip 2015 card given the slow trickle of performance since 290X/780Ti.
http://www.computerbase.de/2014-09/geforce-gtx-980-970-test-sli-nvidia/6/

 

[antihelten]

So where's the cooler?

I don't know, but maybe the guy who made this picture can tell you, seeing as this 295x2 is also missing it's cooler:

All joking aside, are you seriously asking me why a clearly photoshopped image is missing bits?

 

[RussianSensation]

Unfortunately, I'm almost certain that image is a photoshop of the 295X2 card

Exactly. The supposedly leaked shroud for a 300 series hybrid-WC has non-tapered/extended metal edge near the end of the power connector and straight lines at the top, not sections of them.

http://www.overclock3d.net/articles/gpu_displays/amd_radeon_r9_390x_cooler_leaked/1

The irony of all of this is that even if R9 300 series ends up fast, runs cool and quiet, the mainstream uninformed gamer might perceive it as "AMD's cards are so hot and loud that they needed water-cooling from blowing up", instead of embracing the superior tech like when Porsche 911 996 switched from air to the now iconic water-cooled engines. I am not going to be surprised by the ignorance though. :whiste:

I mean right now people are already freaking out about a 300W TDP rating when an R9 290X had an unofficial 300W TDP rating itself and both 780Ti and 290X used a similar amount of power. Again, if 250-300W of real world power usage is a deal breaker, there will be 980/380 mid-range style cards with 175-200W of power usage. Flagship cards are about pushing limits like exotic sports cars. Who complains about the fuel economy of a McLaren P1, LaFerrari or Porsche 918? Really?

The interesting aspect is Asetek won a contract with an OEM for $2-4 million but even if we take $4 million revenue as the high end and assign just $15 BOM for the Asetek cooler, we get just 266,667 R9 390 videocards. Considering NV already sold 1 million GTX970/980 cards, how does AMD plan on meeting the demand with such low manufacturing volume exactly? This one aspect I can't reconcile unless the $2-4 million Asetek contract is just for the first batch/one quarter of cards and AMD will renew the agreement thereafter. It's one thing to make 250,000 Hybrid-WC cards but that's a drop in the bucket for the demand AMD will have if 390X is 50% faster than an R9 290X. Could be a very risky decision that might backfire as prices will rise due to shortages aka bitcoin mining times. :hmm:

 

[Silverforce11]

Have you considered the possibility the top-tier, R380X = hybrid water while the R380 = aircooled (custom AIB designs!)?

That would instantly differentiate the product, giving more incentive for people to buy the X variant, paying the premium.

 

[therealnickdanger]

Dp 1.3?

 

[RussianSensation]

Have you considered the possibility the top-tier, R380X = hybrid water while the R380 = aircooled (custom AIB designs!)?

That would instantly differentiate the product, giving more incentive for people to buy the X variant, paying the premium.

No, I haven't. That would certainly be a very interesting way to create incentive for people to pay extra for the X version considering so many basically skip the X, get a non-X and overclock/unlock.

 

[n0x1ous]

Have you considered the possibility the top-tier, R380X = hybrid water while the R380 = aircooled (custom AIB designs!)?

That would instantly differentiate the product, giving more incentive for people to buy the X variant, paying the premium.

this seems quite plausible to me. For me, that would push me to get the XT over the Pro

 

[Creig]

I mean right now people are already freaking out about a 300W TDP rating when an R9 290X had an unofficial 300W TDP rating itself and both 780Ti and 290X used a similar amount of power. Again, if 250-300W of real world power usage is a deal breaker, there will be 980/380 mid-range style cards with 175-200W of power usage. Flagship cards are about pushing limits like exotic sports cars. Who complains about the fuel economy of a McLaren P1, LaFerrari or Porsche 918? Really?

Four years ago, the Nvidia GTX 480 sucked down nearly 260 watts at peak draw. And yet the national power grid was left intact, contrary to the predictions of some people regarding the upcoming release of the 380X.

 

[raghu78]

The interesting aspect is Asetek won a contract with an OEM for $2-4 million but even if we take $4 million revenue as the high end and assign just $15 BOM for the Asetek cooler, we get just 266,667 R9 390 videocards. Considering NV already sold 1 million GTX970/980 cards, how does AMD plan on meeting the demand with such low manufacturing volume exactly? This one aspect I can't reconcile unless the $2-4 million Asetek contract is just for the first batch/one quarter of cards and AMD will renew the agreement thereafter. It's one thing to make 250,000 Hybrid-WC cards but that's a drop in the bucket for the demand AMD will have if 390X is 50% faster than an R9 290X. Could be a very risky decision that might backfire as prices will rise due to shortages aka bitcoin mining times. :hmm:

Do you know how many GTX 970 cards were sold in that million. Given the outstanding price perf of GTX 970 and the fact that the big unit volumes in GPU market are closer to the USD 250 - 300 price point, I would not be surprised if its 4:1 or 3:1 in favour of GTX 970. In that case the number of GTX 980 cards sold is around 200,000 - 250,000. btw if the R9 390X is around 60 - 65% faster than R9 290X it will take the GPU crown and AMD should be able to price it closer to USD 700. As prices go higher volumes will be lower. We know the R9 390 will sell more volume and sell at lower price. That card could come out of the gate with AIB air cooled designs.

Also you forget that the first batch might be AMD reference design and hybrid cooled. Thats how it has been in the past. AMD will definitely allow AIBs to come out quickly with custom versions. These AIBs will push the envelope when it comes to cooling techniques if they know the GPU will have the single GPU crown and they can improve their margins by selling value added designs with more powerful PCBs and improved cooling. Prices will go even higher on the Vapor-X and Lightning versions pushing upward to USD 800. We don't know if those too will be hybrid cooled or air cooled.

 

[Piotrsama]

Hello guys,

I'm the editor in chief of SweClockers and I feel obliged to respond since there seem to be a lot of stuff lost in translation. The news post is about the launch in late Q2, probably somewhere around Computex.

The rest is just a quick recap about the rumors floating around the net. Some more credible than others.

Best regards,
Andreas Dimestam

In case anybody missed that post.
Can we edit OP/thread title to reflect that, so we don't create false expectations?

 

[destrekor]

Exactly. The supposedly leaked shroud for a 300 series hybrid-WC has non-tapered/extended metal edge near the end of the power connector and straight lines at the top, not sections of them.

http://www.overclock3d.net/articles/gpu_displays/amd_radeon_r9_390x_cooler_leaked/1

We can even ignore those differences, and zoom in right on the obvious marks of a badly-made photoshop: look at the middle section, where the fan was on the 295X2 shroud. It's all wavy and distorted.
'shopped. Badly done, at that.