Good info for all so I will just bring it here for everyone, guess we should start with AMD and then some info on Nvidia. It looks like AMD will strike first with the release next month of their new line up, the R-200 series (not the 9000 series everyone expected as the 8000 series was reserved for mobile). Then Nvidia will answer with releases of more 700 series to be followed in Feb-Mar of 2014 with the 800 series release.
The graphics power of today's cards is truly amazing compared to last year at this time, and in a few months both companies will bring things to a new level in performance. I think it is going to be a pretty exciting time in the graphics world and perhaps make high end cards affordable to most everyone. I am hoping mid-high class cards like the 7970, 760, 770 and 780 will drop in price/performance ratios to accommodate the new hardware being offered.
FROM VIDEOCARDZ: AMD
The AMD Radeon 9k series are just around the corner. Few weeks ago we told you that the new series will launch in October. Since then we did not have any proof. That’s until now. Most of you probably know that Hawaii GPU will be shown by AMD in late September. Of course this will only be the official announcement. The new retail models will hit the shelves in mid October according to our latest information.
Digitimes already confirmed that the mass shipment of the Hawaii-based graphics cards will commence in October.
As AMD is set to announce its next-generation high-end GPU codenamed Hawaii, graphics card players including Asustek Computer, Micro-Star International (MSI), Sapphire and PowerColor are expected to start mass shipping related products in October, according to sources from the upstream supply chain.
My source is slightly different. The picture you are going to see is from the headquarters of one of the main AMD AIB partner. Of course I cannot tell you which one, because this could get them into trouble. The funny thing about this is that if they ask me to remove it, they would confirm it at the same time. The photograph is not very straightforward. In fact it’s quite fuzzy and unclear what is shows. I will give you my best bet, but of course you are welcome to add your thoughts right below this post.
So let’s summarize what it probably shows:
Two GPU variants XT and Pro (this could be either Hawaii or other GPU)
Three memory configurations: 8x DDR5, 4x DDR5 and 8x DDR3
Maximum memory capacity of 4GB
Mid-October launch confirmation
The 13xxxxxx labels could potentially mean internal PCB model markings
Some PCB models could use both XT and PRO GPUs
The XT/PRO+{number1}:{number2} label could mean the overclocking level of the memory and the GPU respectively
Apparently one model (probably not the only one) will have two fans, but they need to order them first
We have two important bits of information. Most of the cards will launch in mid October, this will include both DDR3 and DDR5 models. I find it hard to believe that Hawaii or any high-end card would use DDR3 memory, so apparently they would launch entry-level or mid-range segment along with Hawaii cards. The other interesting information is that the manufacturer is planning 4GB cards. At this point we don’t know if this number is referring to Hawaii or other GPU, that’s because we don’t see the whole board, thus we might be missing some information.
So far I have not seen anything that would confirm the new R9-Dxxx naming scheme. That said, I will will continue to use both names until we confirm it.
That’s all for now, stay tuned for new leaks.
NEW UPDATE:
We might have just learned the final name of the new Radeon Volcanic Islands series. With the data provided by our friends from techPowerUp, we can conclude a list of the upcoming Radeon graphics cards.
Although this information was already confirmed by three different sources I cannot guarantee that AMD will change their plans right before the launch, thus please take it with a grain of salt.
AMD Radeon R-200 Series
When I first wrote about the new Radeon naming I wasn’t really sure if that was true. All we had was an article from Semi Accurate and speculation. Today I can tell you more. Apparently AMD made a survey few months ago asking how to name their next cards. One of the possibilities was Radeon Rx 200. According to the latest information this is in fact the final naming of the new cards. The high-end models will probably end up with R9 prefix, other segments will be named under R8, R7, R6 and R5 labels respectively. The Rx symbol will then be followed by a model number. The next series will start from 200, then move to 300 and so on. So yes, it’s pretty much what we’ve expected, only without the Desktop part.
I will give you an example for Radeon HD 7990 successor, supposedly based on dual Hawaii GPU, also known as Tonga. The AMD Radeon R9 290, as it’s believed to be named, is plain and simple. Things get little confusing if we move to Radeon HD 7970 successor. At this time we don’t know if it’s going to be named R9 285 or 280. If we move down the list to HD 7870 successor we will end up with R8 275. Of course this is just the speculation, all we know for sure is that AMD itself suggested R9 290, so we treat it as a starting point.
With the new naming schema, AMD can position their products in more reasonable way, with a smooth transition from high-end to entry-level segments. That’s important since some leaks suggested few Hawaii variants. If that’s the case, then we shall expect cards named from R9-290 to 270 based on on the same processor, only with disabled units and less memory modules (this would also explain quite a few memory configurations in earlier leak).
Furthermore we should also mention that Radeon R-200 series will include many rebranded GPUs. Starting from Cape Verde and Pitcairn and Bonaire.
AMD Radeon R-M200 Series
AMD is also working on mobile graphics cards. It was earlier revealed by an entry in Catalyst files that mobile parts will receive the M prefix right in front of the model number. So far we know about Radeon R5 M200, most likely based on Sun GPU.
Of course that’s not the end of R-200 series. The new naming will most likely be adopted for integrated graphics, expect R-D200 or R-I200 as well.
I will keep digging into this story, if something new pops up I will update this post or make a new entry.
Source: techPowerUP
NOW ABOUT NVIDIA:
New graphics cards this year
New models would probably include more GK104/106 rebrands. NVIDIA could launch more GK208 entry-level models as well. We are probably expecting more Titanium cards with boost mode enabled. With GeForce 600 series NVIDIA launched three x50 cards, I expect 700 series to be no different.
GeForce GTX 790 in few months?
However this is not all. NVIDIA is also working on a new high-end model. At this point I don’t know if this is the so-called TITAN ULTRA with fully enabled GK110 GPU or, more likely, GeForce GTX 790 with two GK110s. Assuming that the full GK110 processor is somewhat dedicated to the professional segment, and NVIDIA made dual-gpu cards a standard, my bet is on GTX 790. You are probably wondering how much would it cost. I’m more than sure that $1000 range won’t be exceeded. This card is made as a response to AMD Hawaii cards which will arrive in two months. Then NVIDIA will most likely (slightly) lower their prices on GK110 graphics cards.
In the end, Santa will have a lot of work this Christmas.
GeForce 800 series “Maxwell” in Q1 2014
Just as I told you in my previous posts, NVIDIA will launch next series sooner than expected. I was told by one of the manufacturers that NVIDIA will release Maxwell series in early 2014. It is expected that GeForce 800 series will arrive in the first quarter of 2014, somewhere between February and March. What it basically means is that the chances for 20nm process are low. Of course NVIDIA could already have first 20nm samples sooner than that, but TSMC will not be ready for mass production till June 2014. Thus, unless my source is wrong the first Maxwell GPUs will not be made in 20nm fabrication process, but if they are expect huge graphics cards shortage and more paper-launches.
A LITTLE INSIGHT:
Quinn asked me if I could post his thoughts about Maxwell architecture being released in the first quarter of 2014, most likely on 28nm process. I encourage you to discuss this topic further in the comments – WhyCry.
If Maxwell is to be hard-launched before 2H2014 it will have to be on 28nm. Nvidia is likely to launch at least a few Maxwell chips before this to compete against new GCN 2.0 chips, coming in October. The reasons why this is happening are diverse. They have to do with the history of GPU and production processes troubles that are making new process nodes worth less and less, a new competitor for TSMC’s cutting edge process, and Nvidia not having a Kepler refresh. Oh, and AMD’s release schedule.
Nvidia and ATI (now AMD) have had a “pattern” when it comes to moving to smaller processes that has continued to the writing of this article to my knowledge. ATI would be the first one onto a node, and would benefit from being able to make smaller GPUs that were as strong as Nvidia’s larger ones. This also meant they dealt with many of the risks and problems that arose in that node, and Nvidia could move onto it once it was safer. Now, to my knowledge, this held true as being what more-or-less happened until the jump to 40nm came.
ATI still leapt ahead of Nvidia, but Nvidia was unable to make a chip that matched ATI’s yields, and the chips were barely faster. That node is when Nvidia had miniscule early yields on its highest end part (about 2%) the GF100, it should also be noted that AMD had yields on its highest parts that were 200-300% over the GF100: about a 5% yield. The node was horrible for yields, AMD fixed its problems quickly, due to having more experience with troublesome nodes. This was the start of the node issues for Nvidia and AMD.
Fast forward a little, TSMC’s 32nm process was canceled due to technologies TSMC was trying for the first time not working, in addition to being power hungry and being scheduled to arrive only shortly before 28nm would. It was delayed multiple times, and eventually canceled it. This caused ATI to have to remake its 6000 series on 40nm, instead of 32nm. A foundry which had the best technology next to Intel (who has the best (large scale) fabrication technology in the world) was failing to make a working node. This is the first time that TSMC had ever done this to my knowledge.
TSMC’s 28nm process was much better than its 40nm was, or its 32nm process would have ended up being. Here is where Nvidia brought up something that is extremely important to why Maxwell is on 28nm to start. In the past, jumping to a new node meant that the cost per transistor would get lower once it was mature, on 20nm, Nvidia does not predict that. Meanwhile, the cost of wafers would start to rise drastically at 20nm. Meaning cost savings gained from smaller chips are lessened. Separately, typically smaller nodes mean it costs more to design chips on those nodes. Starting with 40 to 28nm (I believe) the cost began to rise drastically. To make this worse, foundries like TSMC generally require their partners (AMD, Nvidia, Qualcomm) help them support the cost of getting the new node working. The price of that has gone through the roof, and is continuing to rise.
The reason that probably contributed most to Nvidia deciding to initially make Maxwell on 28nm was due to Apple becoming a fourth player who wanted TSMC’s cutting edge process nodes — Apple — the company that knows what it wants, demands it, and has the money to ensure it gets it
Normally, Nvidia has a refresh, Fermi version one was GF100-GF109. Fermi version two, faster, lower power, and better yielding, was under the badge of GF110-119. Nvidia either did not plan to refresh Kepler, canceled the refresh because it was not a large improvement over Kepler version one. This leads to a major problem, AMD’s GCN2.0, coming out in October-ish, over half a year before 20nm could even launch.
Nvidia needs to remain competitive with AMD, and well GK110 might do well in the high end, the lower end is more valuable. Nvidia could either give up the lower-end by keeping Kepler, lower its margins by selling larger Kepler chips against smaller GCN2.0 chips, or release Maxwell on 28nm.
If Nvidia had not made Maxwell for 20nm, there is a large chance that AMD not only would have had the 28nm GCN 2.0 chips have no real competition, but also, based on history, would launch a 20nm GCN2.0 part before Nvidia had even revealed Maxwell.
According to VC sources, Maxwell will not be as drastic a jump as Tesla to Fermi was, or as Fermi to Kepler was. Mostly it would be updating technology and adding new ways for data to be managed, such as hUMA. Nvidia almost may just improve the CUDA cores in the SMXs, or add more too each SMX. There will be no chip like the GK110 coming on the 28nm node, the Maxwell Refresh onto 20nm will bring the monster chip people are waiting for.
To summarize, Nvidia is probably going to make at least some 28nm Maxwell chips because they don’t know when they could launch 20nm Maxwell in large numbers, they don’t want to launch their new architecture over half a year after AMD launched theirs, and they would be caught in a position of having no speed bump against a massive 20%+ speed bump. Maxwell also ensures that Nvidia won’t fall behind in data management, which is a large part of GCN 2.0. Nvidia is launching 28nm to remain competitive both in power, and in ways to manage memory.
Source: SemiAccurate, CDRinfo, Bit-tech