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

[RussianSensation]

Some points not mentioned:

1) Gamers always needed to wait another 3-6 months for the best flagship card designs such as Galaxy HOF, MSI Lightning, EVGA Classified and Asus Matrix. That means even for air cooled solutions, enthusiasts who wanted the best cards couldn't get them at launch. That means even air cooling brings with it certain compromises.

2) We don't know the launch dates of GM200 vs. flagship 300 series. Everything is just a rumour at this point.

3) IMO there will be more important differentiating factors than AIO vs. air when choosing GM200 vs. 380X/390X (whatever it will be named):

- 4GB vs. 6GB of vram
- 4K gaming performance
- CF vs. SLI scaling and smoothness
- cool features (H.265 support, DP1.3, ZeroCore power, etc)
- overclocking headroom

If GM200 wins at 4K and SLI is smoother, a cooler and quieter 300 card won't be chosen by enthusiasts. The cooling type is just 1 aspect of many for the cards.

For example, almost no one here talks about the ZeroCore feature of AMD cards but they talk about NV cards turning their fans off. With ZeroCore, the 2nd and 3rd card go into hibernation mode where they use something like 3-5W of power.

There are also GameWorks titles. If someone mostly plays those games, well the choice is more towards NV cards.

Trying to pick a winner based on the cooling solution overlooks way too many other important factors. I realize price doesn't matter to those who want the best but if one card is $550-600 and the other $700-800, that starts to matter to some people going dual or triple cards. With the former setup you could get water blocks for "free" so to speak.

We also need to see HardOCP's analysis of SLI. GM200 is even faster than 980 but more than once HardOCP commented how XDMA feels smoother. Sounds to me if both setups were to provide similar FPS, CF would feel smoother.

 

[Silverforce11]

We also need to see HardOCP's analysis of SLI. GM200 is even faster than 980 but more than once HardOCP commented how XDMA feels smoother. Sounds to me if both setups were to provide similar FPS, CF would feel smoother.

[H]'s analysis is not worth much these days. The reason is their small game selection and mostly GameWorks title, particular Watch Dogs & Far Cry 4. You can't draw conclusions when half the games are broken.

I hope they expand their selection and branch out to different genres too.

 

[dacostafilipe]

With memory not needing PCB space anymore, this could lead to really short cards. Would be awesome for people with small cases (like myself).

 

[RussianSensation]

I hope they expand their selection and branch out to different genres too.

Same. There are lots of new 2015 PC games they could use to expand their test suite to 10-12 games.

- Evolve
- Dying Light
- BF Hardline
- The Witcher 3
- Project CARS
- GTA V
- Just Cause 3
- The Division
- AC Victory
- Batman AK

Can easily have a 10-game round-up by end of 2015.

I feel if they did, they would have a stronger shot at becoming THE premium go to site for high-end PC gaming (i.e., 1440P/4K/multi-monitor). I would guess that most gamers with GM200 SLI or similar will not be gaming on 1080P monitors.

 

[Black Octagon]

The other problem with [H] is that they don't do frame time analysis. Just FPS plus their subjective comments

 

[n0x1ous]

Anyone think that the AIB's might not be happy with AMD for going CLC reference as it would make their aftermarket designs LESS desirable when normally the aftermarket designs have been more desirable?

 

[blastingcap]

Anyone think that the AIB's might not be happy with AMD for going CLC reference as it would make their aftermarket designs LESS desirable when normally the aftermarket designs have been more desirable?

Aftermarket has not always been more desirable due to AIB costcutting that goes on underneath the hood, though in recent years that has gotten better. Even XFX learned its lesson apparently... gamers will not tolerate crap like the cooler they used on the 79xx and the long-term harm to XFX's brand is likely more than the short-term cost-cutting is worth.

Anyway, I will not let water in my system, period. I think many feel the same way, but air is usually cheaper. So I would not worry about AIBs, they will get plenty of business using cheaper air coolers.

 

[wand3r3r]

Aftermarket has not always been more desirable due to AIB costcutting that goes on underneath the hood, though in recent years that has gotten better. Even XFX learned its lesson apparently... gamers will not tolerate crap like the cooler they used on the 79xx and the long-term harm to XFX's brand is likely more than the short-term cost-cutting is worth.

Anyway, I will not let water in my system, period. I think many feel the same way, but air is usually cheaper. So I would not worry about AIBs, they will get plenty of business using cheaper air coolers.

For the first part, it's good that consumers can affect corner cutting like that. AIB cutdowns, the 290x reference cooler, the 970 lies, (price hikes i.e. $3k titan-z) can all be avoided in the future by consumers speaking through their wallets.

As for the second part, I've debated switching from the best air cooler to an AIO cpu cooler simply to gain not only all the extra room, but also to get a different airflow situation in my case. I haven't simply since both cost quite a bit and there is no gain to an in place switch. I suspect that my next build will include a nice quiet and beefy AIO cpu cooler. Either way, for those that are opposed to it, there will certainly be traditional fan cooled models.

 

[RussianSensation]

Anyone think that the AIB's might not be happy with AMD for going CLC reference as it would make their aftermarket designs LESS desirable when normally the aftermarket designs have been more desirable?

Another reason why AIO CLC is so amazing as a reference design -- it will force AIBs to use higher quality parts, and continue to improve their coolers to make their cards more attractive. How people don't see this is remarkable. This is AMD providing more competition, which means we as consumers win. As Blastingcap noted AIB's can't roll-over same heatsink and cheap parts designs. They better step up their game unless they want to become the next Albatros, Chaintech, BFG, etc.

Look at the evolution of AIB cards. EVGA is now near bottom of the barrel for quality parts and cooling, while ASUS, MSI and Gigabyte stepped up their game! You can't become complacent. Even Sapphire continues to improve their Vapor-X and Tri-X designs.

 

[destrekor]

By the way, did i mention that there will be cards from partners which will be air-cooled? Yes i have, repeatedly over several pages now. Honestly i'm baffled that why people who don't want an AIO solution, can't go ahead and buy a card with an air-cooled solution like vapor-X or something else, which will obviously come and will be more than adequate. Instead the "open minded" lot continue to rant on about reference card, which they would most likely have never bought in the first place.

Can you not see that I have, almost every time you have said this, responded by saying that I recognize AIBs will create open-air designs? I'm not ignoring it, I'm addressing the idea that CLC's are better. Everyone is all up in arms about how CLCs are better, 100% of the time, no ifs ands or buts, and I'm simply stating, y'all are wrong. They CAN be better, but it is not universal. Especially once you factor in individual situations.

 

[n0x1ous]

Another reason why AIO CLC is so amazing as a reference design -- it will force AIBs to use higher quality parts, and continue to improve their coolers to make their cards more attractive. How people don't see this is remarkable. This is AMD providing more competition, which means we as consumers win. As Blastingcap noted AIB's can't roll-over same heatsink and cheap parts designs. They better step up their game unless they want to become the next Albatros, Chaintech, BFG, etc.

Look at the evolution of AIB cards. EVGA is now near bottom of the barrel for quality parts and cooling, while ASUS, MSI and Gigabyte stepped up their game! You can't become complacent. Even Sapphire continues to improve their Vapor-X and Tri-X designs.

Yeah - EVGA certainly has fallen from grace the last few gens. I really hope AMD does the CLC reference. It will totally negate what happened with reviewers using 290/x reference for the basis of all comparisons. Plus its just a better solution period.

 

[destrekor]

Another reason why AIO CLC is so amazing as a reference design -- it will force AIBs to use higher quality parts, and continue to improve their coolers to make their cards more attractive. How people don't see this is remarkable. This is AMD providing more competition, which means we as consumers win. As Blastingcap noted AIB's can't roll-over same heatsink and cheap parts designs. They better step up their game unless they want to become the next Albatros, Chaintech, BFG, etc.

Look at the evolution of AIB cards. EVGA is now near bottom of the barrel for quality parts and cooling, while ASUS, MSI and Gigabyte stepped up their game! You can't become complacent. Even Sapphire continues to improve their Vapor-X and Tri-X designs.

The only fear is though, how many non-reference or based-on-reference solutions will AIBs create? The ultimate reason for custom coolers has always been more overclocking headroom with lower noise and heat compared to the reference cooler design. If the reference design tolerates all of that and more, there are less reasons to justify the R&D for a new cooler.

I think we'll see them, but I don't know how much focus and effort AIBs will put forth on such designs. Ultimately, I think the only reason AIBs will create any other designs is simply to have options for the consumer, but, I think we'll see fewer options and perhaps even the best OC cards on reference or reference-style only. I hope I'm wrong.

 

[DeathReborn]

I could only have 1 CLC based GPU in my current gaming system. I have a Lian Li PC-A70S which runs 2x power supplies for better load balancing, a 240mm RAD for CPU cooling and has only 1 other usable (for a rad) 120mm exhaust. Intakes are 4x 120mm front fans with the HDD cage between them, 240mm Radiator on the top of the case & 1 of the 120mm rear slots used fort the 2nd PSU.

Currently this copes with R9 290 CF 1150/1400 & a 4770K @ 4.8GHz so I would either wait for a decent 3rd party 380X, 380X2 (or whatever they call it) or go with GM200. I am very happy with my case and am not going to change it for extra Rad capacity.

 

[RussianSensation]

Sell both of those PSU and just get a 1000W SeaSonic or similar. Problem solved. Running 2 PSUs is extremely strange unless no single PSU in the world can sustain your load. Considering we have 1500-2000W PSUs, I find this impossible to believe.

 

[tential]

Same. There are lots of new 2015 PC games they could use to expand their test suite to 10-12 games.

- Evolve
- Dying Light
- BF Hardline
- The Witcher 3
- Project CARS
- GTA V
- Just Cause 3
- The Division
- AC Victory
- Batman AK

Can easily have a 10-game round-up by end of 2015.

I feel if they did, they would have a stronger shot at becoming THE premium go to site for high-end PC gaming (i.e., 1440P/4K/multi-monitor). I would guess that most gamers with GM200 SLI or similar will not be gaming on 1080P monitors.

Game roundups are supposed to include new and old games so you can get a sense of how GPUs will perform now and how they will perform on past games.

All new games would distort the picture, most likely in NV's favor as well since they tend to have better launch day drivers for games.

Also new games generally have tons of launch issues so using only new game this year in reviews would be a testament to how good the games were made and not how good the GPUs are.

 

[iiiankiii]

The only fear is though, how many non-reference or based-on-reference solutions will AIBs create? The ultimate reason for custom coolers has always been more overclocking headroom with lower noise and heat compared to the reference cooler design. If the reference design tolerates all of that and more, there are less reasons to justify the R&D for a new cooler.

I think we'll see them, but I don't know how much focus and effort AIBs will put forth on such designs. Ultimately, I think the only reason AIBs will create any other designs is simply to have options for the consumer, but, I think we'll see fewer options and perhaps even the best OC cards on reference or reference-style only. I hope I'm wrong.

Like always, they will create both reference designs and custom designs. Just look around. There are short cards, long cards, open air, blower styles, etc. In fact, there are even custom water cool ready GPUs in the market right now. Don't worry, there will be plenty of options out there because the market wants your money! It'll try its hardest to get it.

 

[RampantAndroid]

It has been demonstrated over and over, that a 290x or a 380x doesn't need water-cooling as a standard fitment, but instead is being provided with one as extra value for reference cards, setting new standards while at it.

You clearly have established only two things:
- you have a preference for air cooled cards
- you don't understand much about efficiency which you talk about, i.e., performance/ watt.

It's a 120mm fan which is what is suggested. Surely there aren't that many mainstream cabinets that can't fit a 120mm fan. I genuinely don't know anyone who doesn't have a space for a 120mm fan, do you?

Wait, we know the exact specs of the 380 now? And you can say they're an absolute? Give me a break.

How many people do you know who want to do CF? How many people do you know that already have a closed loop water CPU cooler? I give up repeating my point. You're clearly ignoring, or skimming.

I'm guessing you have a 780Ti since it's about on par with a 970/290X. If that is the case, a real upgrade is going to require a higher TDP than the 980 has.

As posted earlier (guessing you missed it) - 2x ASUS 670s.

 

[tential]

Aftermarket has not always been more desirable due to AIB costcutting that goes on underneath the hood, though in recent years that has gotten better. Even XFX learned its lesson apparently... gamers will not tolerate crap like the cooler they used on the 79xx and the long-term harm to XFX's brand is likely more than the short-term cost-cutting is worth.

Anyway, I will not let water in my system, period. I think many feel the same way, but air is usually cheaper. So I would not worry about AIBs, they will get plenty of business using cheaper air coolers.

For the target market of this graphics card (Which will likely debut at $500+ given its performance level) you're in a very very small minority of high end gamers who don't like water cooling.
The loss of people who "Won't ever let water near their rig" is worth the reward of many high end gamers who aren't necessarily hardware enthusiasts who want a simple way to get a water cooled GPU.

Actually I think a lot of people feel like I do about water cooling. They want a water cooled GPU/CPU, but they don't want to do any type of modification at all (things you guys find easy). I've never considered water cooling because it was an aftermarket modification I had to do to my GPU that I didn't ever feel like doing.

A water cooled GPU CLC shipped as the reference design, for me that's the first time I'll consider water cooling and for many the feeling will be the same.

So ya, they'll lose people like yourself, but they'll gain lots of others who buy high end cards and want an easy simple way to get water cooling with little to no effort involved.

 

[garagisti]

Can you not see that I have, almost every time you have said this, responded by saying that I recognize AIBs will create open-air designs? I'm not ignoring it, I'm addressing the idea that CLC's are better. Everyone is all up in arms about how CLCs are better, 100% of the time, no ifs ands or buts, and I'm simply stating, y'all are wrong. They CAN be better, but it is not universal. Especially once you factor in individual situations.

That's the problem right there. You are not debating whether AIO is better than air scientifically, but instead are speaking of individual cases who may want air cooled kit owing to limitations and perceptions. Then again, before now you left it so loosely worded which read as AIO may not be a good thing for reference. Fact is that AIO as reference is just plain superior a solution for cooling, and while it may not suit you, it doesn't make it inferior.

 

[garagisti]

Wait, we know the exact specs of the 380 now? And you can say they're an absolute? Give me a break.

How many people do you know who want to do CF? How many people do you know that already have a closed loop water CPU cooler? I give up repeating my point. You're clearly ignoring, or skimming.



As posted earlier (guessing you missed it) - 2x ASUS 670s.

On the contrary, it is you who has been skimming and ignoring several posts about efficiency which you have been drumming up. By the way, the rumored performance is what we were talking about all this while, and it didn't stop you from bashing the card over several pages, many lines where you complained how AIO is needed and that's bad, despite suggestions that it's only 10w more than 290x in the same rumors. Since your argument has repeatedly been proven to have no merit what so ever, now you say that it's all rumours and we should not have hang ups about it. Did I get that right?

 

[destrekor]

That's the problem right there. You are not debating whether AIO is better than air scientifically, but instead are speaking of individual cases who may want air cooled kit owing to limitations and perceptions. Then again, before now you left it so loosely worded which read as AIO may not be a good thing for reference. Fact is that AIO as reference is just plain superior a solution for cooling, and while it may not suit you, but it doesn't make it inferior.

It's a superior solution, but again the point I have argued is that, it's not the best, due to the possible issues, let alone potential case-fit issues. People may have room for one, not two. People have removed drive cages and turned front intake into exhaust, which is two problems immediately for many people (they need the cage, and front exhaust = inefficient airflow, which may compromise other cooling).

And a good reason to worry with AIO/CLC as reference: everyone argues that both CLC and Air will be available, but this is simply guessing, because all other reference (air) has been supplanted by custom solutions. Nobody changed the CLC on the 295X2, did they? And others have added water as a solution, due to greater efficiency. Custom cards, as RS alluded to, will also be delayed. Now, how motivated will AIBs be to release a custom cooling solution is is inferior to the reference cooler? I highly, highly doubt any custom air design will be cooler and quieter and offer more OC headroom than the CLC. So, my bet is that the only custom air designs are simply to appease those who can't fit the CLC cooler, and those cards won't likely receive nearly as many beneficial aftermarket tweaks as their main designs will, which will either have the reference-brand cooler or a custom-design that is still a CLC.

I would love to be wrong about this, really, I would. And it is useless to argue now, because AMD will be making up their mind however they see fit, and we'll just have to wait and see how the market and AIBs handle that situation.

 

[sandorski]

AIBs might keep the CLC cooler and differentiate themselves more with asthetics. Thinner shrouds or just interesting designs of the shrouds. It could lead to some interesting results.

 

[RampantAndroid]

On the contrary, it is you who has been skimming and ignoring several posts about efficiency which you have been drumming up. By the way, the rumored performance is what we were talking about all this while, and it didn't stop you from bashing the card over several pages, many lines where you complained how AIO is needed and that's bad, despite suggestions that it's only 10w more than 290x in the same rumors. Since your argument has repeatedly been proven to have no merit what so ever, now you say that it's all rumours and we should not have hang ups about it. Did I get that right?

For the *last time* I did not bash this card. I made a generic statement that if a card required a closed water loop to function, the company making it was doing it wrong. I have NEVER (I cannot stress this enough) NEVER said "the 380/390 is crap."

 

[xthetenth]

I very strongly doubt that companies which have invested a lot into their aftermarket coolers would suddenly drop them because those coolers have an even greater appeal to part of their market, which can't go reference.

I also personally feel that the only card to require a closed loop cooler was not in fact doing it wrong but was offering a very compelling product with the first two GPUs on one slot card to not take a clock speed hit in recent memory.

 

[Headfoot]

For the *last time* I did not bash this card. I made a generic statement that if a card required a closed water loop to function, the company making it was doing it wrong. I have NEVER (I cannot stress this enough) NEVER said "the 380/390 is crap."

If a card needs an air cooler it's doing it wrong.

Why not just have bare chips? Just make cards that use less power. No moving parts, no extra cost of the fan. Win win right?