AMD CPU speculation... and expert conjecture

[Cazalan]

Compared to Trinity's die, it's only slightly smaller, but I'm itching to see what it's made up of. Now that we have Steamroller on its way, this thread will grow even larger!

Based on some rough math comparing the die shots I came up with 1.42bil transistors which is only 120mil more than Trinity. Not enough for eDRAM.

96.8% smaller than Trinity where a straight shrink would be 87.5%. 9.3% larger than the node shrink. 1.303B to 1.424B transistors.

They could go the GT3e route and add it on a new larger package.

 

[esrever]

why would you want to see per clock performance of richland to trinity? all the work they did was on the power management with small adjustments to turbo. The die is exactly the same. Even the ddr3 2133 support comes as a bios update that also works on trinity chips. The chip's advantages really just lie with being higher clocked within the same power envelope. The overclock is also amazing.

Kaveri is likely very basic for what it offers. Not going to get anywhere near 7800 performance. By the looks of it, it will offer likely 384 GCN cores, 4 steamroller cores and probably support higher clocked ddr3. I expect that alone to give a 40-50% increase in performance on the IGP. Considering kabini is about 1/2 as fast as trinity and Kaveri effectively triples the performance of kabini with these specs. The HSA enhancements will be the biggest things. Just hope to see some software support.

 

[JAYDEEJOHN]

Some low hanging fruit picked, more to come

 

[Cazalan]

Kaveri is likely very basic for what it offers. Not going to get anywhere near 7800 performance. By the looks of it, it will offer likely 384 GCN cores, 4 steamroller cores and probably support higher clocked ddr3. I expect that alone to give a 40-50% increase in performance on the IGP. Considering kabini is about 1/2 as fast as trinity and Kaveri effectively triples the performance of kabini with these specs. The HSA enhancements will be the biggest things. Just hope to see some software support.

They could bump it to 7 CU for 448 GCN cores. Might put it on par with a 7750 which isn't bad at all. Getting to 7800 performance would require a radical change like the PS4. Just not feasible for 28nm with bulky SR cores too.

 

[sarinaide]

I would target HD7770 performance, then its truely mainstream low end and 1080 ready. A while ago they said AMD would double the SP of Trinity that's a hair over 750 Stream processors for the top end part. current parts are 8 ROP's and 48 TMU's may need that and the memory bus bumped.

 

[GOM3RPLY3R]

Well, the first 64 bit technology intel used was a system that flopped, they eventually just adopted for use all the AMD x86-64 instruction sets that AMD introduced (AMD was first btw). I cannot even recall what it was they called it...

I have a feeling this will go the same way...the players involved in HSA are too big and too many for Intel to overcome all of them. Much like Nvidia's CUDA is battling OpenCL, I think time will be on the foundation's side. There are many more adopters of OpenCL than there are for CUDA (scientific community is about all there is for CUDA adopters outside of individual companies).

I'm just going to say, its a little cocky to say that they were first. Yes they are, but trying to make that a point to try to win, is not right.

Also with CUDAand OpenCL, OpenCL is not as much as used as Dx11 and Dx10, which is what CUDA is better adapted for than OCL. AMD in a similar way is better with OpenCL than Dx. Food for Thought.

1.) It's not cocky...it's truth.

2.) OpenCL is not a competitor to DirectX in anyway shape or form...that's OpenGL.

3.) Nvidia and AMD both support DirectX11 and OpenGL 4.2.

4.) OpenCL is something completely different...and while NVidia supports this some, AMD is big on OpenCL. NVidia's counterpart is CUDA, which must be programmed for specifically to work on NVidia GPUs as it's proprietary. Any GPU can run OpenCL as it's an open standard...that's why CUDA will fail.

5.) OpenCL and CUDA are about using the GPU to do parallel processing by running compute functions on GPUs.

6.) DX and OpenGL are about 3D hardware acceleration and graphics APIs, as well as physics computations and other media tasks.

You might want to do some homework before you run around calling someone "cocky".

I have done my homework and understand what all this is. AMD does have the upper hand with OpenCL, and everyone knows that. However, how will CUDA fail if many corporations such as Bohemia Interactice, DICE, and many other are expanding with it? Most of the ArmA series is based off of Nvidia Physx which is greatly boosted with the more CUDA cores that you have.

Even so CUDA does still perform at a great level. The only advantage that AMD has is:
- obviously with video and photo editing
- cheaper, so its better for tighter budget people
- and games that use a majority of the OpenCL system

With all this, I may also add that with the architecture of games that are OpenCL based, isn't really catching up to any of the other programs. Yeah the game can sometimes run faster with it, but the quality to performance ratio is lacking on the quality side with it.

 

[Cazalan]

I would target HD7770 performance, then its truely mainstream low end and 1080 ready. A while ago they said AMD would double the SP of Trinity that's a hair over 750 Stream processors for the top end part. current parts are 8 ROP's and 48 TMU's may need that and the memory bus bumped.

To double the SPs they'd have to jump to like 2Billion transistors (approx 125mil Transistors per CU). Couple that with the power draw of SR you'd have over a 200W CPU. Not likely without a hefty downclock of the CPU cores.

You're seeing why for PS4 they had to drop in the much smaller Jaguar cores to offset the 18 CU.

 

[sarinaide]

I don't see anything wrong with dropping clock speed, drop to 3Ghz-3.3ghz boost that alone is worth 60-80w.

http://wccftech.com/amd-fm2-socket-kaveri-apu-compatible-trinity-richland/

Pin layout looks different, and that motherboard looks really odd, but it is a Biostar.

 

[esrever]

Kaveri is likely very basic for what it offers. Not going to get anywhere near 7800 performance. By the looks of it, it will offer likely 384 GCN cores, 4 steamroller cores and probably support higher clocked ddr3. I expect that alone to give a 40-50% increase in performance on the IGP. Considering kabini is about 1/2 as fast as trinity and Kaveri effectively triples the performance of kabini with these specs. The HSA enhancements will be the biggest things. Just hope to see some software support.

They could bump it to 7 CU for 448 GCN cores. Might put it on par with a 7750 which isn't bad at all. Getting to 7800 performance would require a radical change like the PS4. Just not feasible for 28nm with bulky SR cores too.

I really doubt this will happen with kaveri, probably will be 8 CU at 20nm for a follow up. APUs are meant to be low power and cheap, even higher end ones 7 CUs are also very odd.

 

[juanrga]

I remember not too long ago AMD expressed desire to turn itself into a gaming company. This may be closer to truth as they announced "Kaveri APU will place a huge emphasis on gaming". From: http://blog.laptopmag.com/amd-reveals-4the-generation-kaveri-apu-at-computex-2013

I think they are basically focusing on gaming now: from tablets to servers.

 

[juanrga]

Richland bench anyone?

http://www.overclock3d.net/reviews/cpu_mainboard/amd_a10-6800k_richland_review/1

That's what I was afraid of. The sky high numbers AMD put out were for synthetic benchmarks. Real world games got 4% improvement if that. Ouch!


EDIT: SA has numbers up too. http://semiaccurate.com/2013/06/04/amd-brings-richland-to-the-desktop-the-a10-6800k/

Reread it VERY carefully and you'll notice something.

http://www.overclock3d.net/reviews/cpu_mainboard/amd_a10-6800k_richland_review/2

4GB Kingston HyperX 1600MHz @ 1333MHz

The review is essentially trash. Trinity requires DDR3-1866 at a minimum, Richland's beefier GPU means it would need DDR3-2133 to really hit it's stride. Running it at 1333 just results in memory limiting the iGPU, mind as well insert a Trinity or Llano at that point in time.

I think that 80% of reviews of a10 trinity that I saw did not use the stock 1866 MHz but always slower. It was a common tactic, specially in Intel vs AMD reviews (where the Intel chips were always using the stock memory). Once I saw one review where a trinity chip was using underclocked memory but the intel chip used overclocked memory.

Huh .... did you read the S/A article? They specifically mentioned that all they had on hand in the lab was DDR3-1833 and that they really wished they had DDR3-2166 for testing. AMD didn't provide them with memory, they used their own.

Am I the only one who thinks that it would be in AMD's best interest to ship the $100 worth of faster RAM to reviewers to get better reviews? I mean, WTH is AMD thinking?

Did you read the posts above? The review? The 'reviewers' had 1600 Ram but clocked it at 1333. In the comment sections they claim had no time to select other speed (LOL). Sent them a superb 2400 kit and all what you want, that they will run again at 1333...

 

[Cazalan]

I don't see anything wrong with dropping clock speed, drop to 3Ghz-3.3ghz boost that alone is worth 60-80w.

http://wccftech.com/amd-fm2-socket-kaveri-apu-compatible-trinity-richland/

Pin layout looks different, and that motherboard looks really odd, but it is a Biostar.

That's an AOI/SFF type layout. Notice no PCIe slots.

 

[griptwister]

I have done my homework and understand what all this is. AMD does have the upper hand with OpenCL, and everyone knows that. However, how will CUDA fail if many corporations such as Bohemia Interactice, DICE, and many other are expanding with it? Most of the ArmA series is based off of Nvidia Physx which is greatly boosted with the more CUDA cores that you have.

Even so CUDA does still perform at a great level. The only advantage that AMD has is:
- obviously with video and photo editing
- cheaper, so its better for tighter budget people
- and games that use a majority of the OpenCL system

With all this, I may also add that with the architecture of games that are OpenCL based, isn't really catching up to any of the other programs. Yeah the game can sometimes run faster with it, but the quality to performance ratio is lacking on the quality side with it.

No, You didn't do your homework. PhysX could run better on a CPU than it would CUDA cores. PhysX is a marketing scheme. I like Nvidia, I'm just sick of all this forced ideas on how gaming should be. If I could afford it, I'd buy a 780 right now, not for PhysX, but because of the performance of the GPU and the capabilities it has with Adobe. Imo, Once Radeon fixes their drivers and makes stronger GPU cores, we're going to see massive improvements across the board and everyone will be neck and neck.

 

[GOM3RPLY3R]

I have done my homework and understand what all this is. AMD does have the upper hand with OpenCL, and everyone knows that. However, how will CUDA fail if many corporations such as Bohemia Interactice, DICE, and many other are expanding with it? Most of the ArmA series is based off of Nvidia Physx which is greatly boosted with the more CUDA cores that you have.

Even so CUDA does still perform at a great level. The only advantage that AMD has is:
- obviously with video and photo editing
- cheaper, so its better for tighter budget people
- and games that use a majority of the OpenCL system

With all this, I may also add that with the architecture of games that are OpenCL based, isn't really catching up to any of the other programs. Yeah the game can sometimes run faster with it, but the quality to performance ratio is lacking on the quality side with it.

No, You didn't do your homework. PhysX could run better on a CPU than it would CUDA cores. PhysX is a marketing scheme. I like Nvidia, I'm just sick of all this forced ideas on how gaming should be. If I could afford it, I'd buy a 780 right now, not for PhysX, but because of the performance of the GPU and the capabilities it has with Adobe. Imo, Once Radeon fixes their drivers and makes stronger GPU cores, we're going to see massive improvements across the board and everyone will be neck and neck.

Once Radeon fixes their drivers and makes stronger GPU cores

I completely agree. Once they do that, then there will be no problem. However, as for the moment, you basically just said Nvidia and their cards are better. I'm just putting that out there.

I would also like to point out: For AMD to make something that is stronger and better than Nvidia's cards, on a Single GPU scale, the price jump would be massive. They would probably be about the same price, if not more expensive. I believe that AMD can do it, its just a matter of how.

Physx may be a marketing scheme, but in the least, it does work great. Even if it may not work better with certain things, the fact that you get Physx and the fact that you are getting a great card, is what matters.

Back to CPUs. I would like to talk about the new A10-6800k. I think this is isanely awesome. For something to take no more than 150 watts and have the power of an i5-3570k and a GTX 670 for about $150 is amazing.

The only thing about this is this:

1) Still produces more heat
2) Requires different boards (for better performance).

I am leaving out the TDP (which is what I mostly don't like AMD for) because it is a fully fledged APU that runs below 150 watts.

The Intel Processors with graphical function can be considered APUs. HOWEVER with something that has GTX 670 level graphics compared to the usual Intel Graphics, is completely incomparable and unfair to Intel. The only thing that may pass is the new Iris graphics, but other than that, the power of the onboard graphics is very well like comparing a GTX Titan to an Nvidia FX 5700, or a Radeon 7970 to a Radeon 2000 series card.

Thus, on a fair ball field, the two CPU's cannot be used. I have to say that in terms of APUs, AMD has really come in far first. I give them a round of applause and 2 thumbs up .

 

[8350rocks]

Well, the first 64 bit technology intel used was a system that flopped, they eventually just adopted for use all the AMD x86-64 instruction sets that AMD introduced (AMD was first btw). I cannot even recall what it was they called it...

I have a feeling this will go the same way...the players involved in HSA are too big and too many for Intel to overcome all of them. Much like Nvidia's CUDA is battling OpenCL, I think time will be on the foundation's side. There are many more adopters of OpenCL than there are for CUDA (scientific community is about all there is for CUDA adopters outside of individual companies).

I'm just going to say, its a little cocky to say that they were first. Yes they are, but trying to make that a point to try to win, is not right.

Also with CUDAand OpenCL, OpenCL is not as much as used as Dx11 and Dx10, which is what CUDA is better adapted for than OCL. AMD in a similar way is better with OpenCL than Dx. Food for Thought.

1.) It's not cocky...it's truth.

2.) OpenCL is not a competitor to DirectX in anyway shape or form...that's OpenGL.

3.) Nvidia and AMD both support DirectX11 and OpenGL 4.2.

4.) OpenCL is something completely different...and while NVidia supports this some, AMD is big on OpenCL. NVidia's counterpart is CUDA, which must be programmed for specifically to work on NVidia GPUs as it's proprietary. Any GPU can run OpenCL as it's an open standard...that's why CUDA will fail.

5.) OpenCL and CUDA are about using the GPU to do parallel processing by running compute functions on GPUs.

6.) DX and OpenGL are about 3D hardware acceleration and graphics APIs, as well as physics computations and other media tasks.

You might want to do some homework before you run around calling someone "cocky".

I have done my homework and understand what all this is. AMD does have the upper hand with OpenCL, and everyone knows that. However, how will CUDA fail if many corporations such as Bohemia Interactice, DICE, and many other are expanding with it? Most of the ArmA series is based off of Nvidia Physx which is greatly boosted with the more CUDA cores that you have.

Even so CUDA does still perform at a great level. The only advantage that AMD has is:
- obviously with video and photo editing
- cheaper, so its better for tighter budget people
- and games that use a majority of the OpenCL system

With all this, I may also add that with the architecture of games that are OpenCL based, isn't really catching up to any of the other programs. Yeah the game can sometimes run faster with it, but the quality to performance ratio is lacking on the quality side with it.

OpenCL will catch on more with each step of HSA integration. Hardware is not yet fully integrated...so how can OpenCL be widespread yet?

Additionally...CUDA is used by some scientific communities...however, notice that AMD has most of the design wins in the Supercomputers where GPGPU compute functions are most thoroughly utilized? You don't see many HPC's running Nvidia...there's a reason for that.

 

[The Q6660 Inside]

I have done my homework and understand what all this is. AMD does have the upper hand with OpenCL, and everyone knows that. However, how will CUDA fail if many corporations such as Bohemia Interactice, DICE, and many other are expanding with it? Most of the ArmA series is based off of Nvidia Physx which is greatly boosted with the more CUDA cores that you have.

Even so CUDA does still perform at a great level. The only advantage that AMD has is:
- obviously with video and photo editing
- cheaper, so its better for tighter budget people
- and games that use a majority of the OpenCL system

With all this, I may also add that with the architecture of games that are OpenCL based, isn't really catching up to any of the other programs. Yeah the game can sometimes run faster with it, but the quality to performance ratio is lacking on the quality side with it.

No, You didn't do your homework. PhysX could run better on a CPU than it would CUDA cores. PhysX is a marketing scheme. I like Nvidia, I'm just sick of all this forced ideas on how gaming should be. If I could afford it, I'd buy a 780 right now, not for PhysX, but because of the performance of the GPU and the capabilities it has with Adobe. Imo, Once Radeon fixes their drivers and makes stronger GPU cores, we're going to see massive improvements across the board and everyone will be neck and neck.

Once Radeon fixes their drivers and makes stronger GPU cores

I completely agree. Once they do that, then there will be no problem. However, as for the moment, you basically just said Nvidia and their cards are better. I'm just putting that out there.

I would also like to point out: For AMD to make something that is stronger and better than Nvidia's cards, on a Single GPU scale, the price jump would be massive. They would probably be about the same price, if not more expensive. I believe that AMD can do it, its just a matter of how.

Physx may be a marketing scheme, but in the least, it does work great. Even if it may not work better with certain things, the fact that you get Physx and the fact that you are getting a great card, is what matters.

Back to CPUs. I would like to talk about the new A10-6800k. I think this is isanely awesome. For something to take no more than 150 watts and have the power of an i5-3570k and a GTX 670 for about $150 is amazing.

The only thing about this is this:

1) Still produces more heat
2) Requires different boards (for better performance).

I am leaving out the TDP (which is what I mostly don't like AMD for) because it is a fully fledged APU that runs below 150 watts.

The 6800K is has GTX 670+3570K power for $150. The Intel graphics are bested by the AMD APUs at identical price ranges. Comparing it is like comparing a GTX Titan to a http:// or a 7970 to a 2900XTX. (Fixed for you.)

GTX 670+3570K. LOLOLOLOLOLOLNO, more like 7750 or 6770 range.

 

[8350rocks]

Spoiler

I have done my homework and understand what all this is. AMD does have the upper hand with OpenCL, and everyone knows that. However, how will CUDA fail if many corporations such as Bohemia Interactice, DICE, and many other are expanding with it? Most of the ArmA series is based off of Nvidia Physx which is greatly boosted with the more CUDA cores that you have.

Even so CUDA does still perform at a great level. The only advantage that AMD has is:
- obviously with video and photo editing
- cheaper, so its better for tighter budget people
- and games that use a majority of the OpenCL system

With all this, I may also add that with the architecture of games that are OpenCL based, isn't really catching up to any of the other programs. Yeah the game can sometimes run faster with it, but the quality to performance ratio is lacking on the quality side with it.

No, You didn't do your homework. PhysX could run better on a CPU than it would CUDA cores. PhysX is a marketing scheme. I like Nvidia, I'm just sick of all this forced ideas on how gaming should be. If I could afford it, I'd buy a 780 right now, not for PhysX, but because of the performance of the GPU and the capabilities it has with Adobe. Imo, Once Radeon fixes their drivers and makes stronger GPU cores, we're going to see massive improvements across the board and everyone will be neck and neck.

Once Radeon fixes their drivers and makes stronger GPU cores

I completely agree. Once they do that, then there will be no problem. However, as for the moment, you basically just said Nvidia and their cards are better. I'm just putting that out there.

I would also like to point out: For AMD to make something that is stronger and better than Nvidia's cards, on a Single GPU scale, the price jump would be massive. They would probably be about the same price, if not more expensive. I believe that AMD can do it, its just a matter of how.

Physx may be a marketing scheme, but in the least, it does work great. Even if it may not work better with certain things, the fact that you get Physx and the fact that you are getting a great card, is what matters.

Back to CPUs. I would like to talk about the new A10-6800k. I think this is isanely awesome. For something to take no more than 150 watts and have the power of an i5-3570k and a GTX 670 for about $150 is amazing.

The only thing about this is this:

1) Still produces more heat
2) Requires different boards (for better performance).

I am leaving out the TDP (which is what I mostly don't like AMD for) because it is a fully fledged APU that runs below 150 watts.

The Intel Processors with graphical function can be considered APUs. HOWEVER[strike] with something that has GTX 670 level graphics compared to the usual Intel Graphics, is completely incomparable and unfair to Intel. The only thing that may pass is the new Iris graphics.[strike]Other than that, the power of the onboard graphics is very well like comparing a GTX Titan to an Nvidia FX 5950, or a Radeon 7970 to a Radeon 2900XTX card.

Thus, on a fair ball field, the two CPU's cannot be used. I have to say that in terms of APUs, AMD has really come in far first. I give them a round of applause and 2 thumbs up .

[/strike][/strike]

GTX 670+3570K. LOLOLOLOLOLOLNO, more like 7750 or 6770 range.

I think he forgot where his bridge was...LOL.

 

[The Q6660 Inside]

Spoiler

I have done my homework and understand what all this is. AMD does have the upper hand with OpenCL, and everyone knows that. However, how will CUDA fail if many corporations such as Bohemia Interactice, DICE, and many other are expanding with it? Most of the ArmA series is based off of Nvidia Physx which is greatly boosted with the more CUDA cores that you have.

Even so CUDA does still perform at a great level. The only advantage that AMD has is:
- obviously with video and photo editing
- cheaper, so its better for tighter budget people
- and games that use a majority of the OpenCL system

With all this, I may also add that with the architecture of games that are OpenCL based, isn't really catching up to any of the other programs. Yeah the game can sometimes run faster with it, but the quality to performance ratio is lacking on the quality side with it.

No, You didn't do your homework. PhysX could run better on a CPU than it would CUDA cores. PhysX is a marketing scheme. I like Nvidia, I'm just sick of all this forced ideas on how gaming should be. If I could afford it, I'd buy a 780 right now, not for PhysX, but because of the performance of the GPU and the capabilities it has with Adobe. Imo, Once Radeon fixes their drivers and makes stronger GPU cores, we're going to see massive improvements across the board and everyone will be neck and neck.

Once Radeon fixes their drivers and makes stronger GPU cores

I completely agree. Once they do that, then there will be no problem. However, as for the moment, you basically just said Nvidia and their cards are better. I'm just putting that out there.

I would also like to point out: For AMD to make something that is stronger and better than Nvidia's cards, on a Single GPU scale, the price jump would be massive. They would probably be about the same price, if not more expensive. I believe that AMD can do it, its just a matter of how.

Physx may be a marketing scheme, but in the least, it does work great. Even if it may not work better with certain things, the fact that you get Physx and the fact that you are getting a great card, is what matters.

Back to CPUs. I would like to talk about the new A10-6800k. I think this is isanely awesome. For something to take no more than 150 watts and have the power of an i5-3570k and a GTX 670 for about $150 is amazing.

The only thing about this is this:

1) Still produces more heat
2) Requires different boards (for better performance).

I am leaving out the TDP (which is what I mostly don't like AMD for) because it is a fully fledged APU that runs below 150 watts.

The Intel Processors with graphical function can be considered APUs. HOWEVER[strike] with something that has GTX 670 level graphics compared to the usual Intel Graphics, is completely incomparable and unfair to Intel. The only thing that may pass is the new Iris graphics.[strike]Other than that, the power of the onboard graphics is very well like comparing a GTX Titan to an Nvidia FX 5950, or a Radeon 7970 to a Radeon 2900XTX card.

Thus, on a fair ball field, the two CPU's cannot be used. I have to say that in terms of APUs, AMD has really come in far first. I give them a round of applause and 2 thumbs up .

[/strike]

GTX 670+3570K. LOLOLOLOLOLOLNO, more like 7750 or 6770 range.

I think he forgot where his bridge was...LOL.

[strike]PUT MY HEATSINK BACK ON![/strike]

 

[GOM3RPLY3R]

Well, the first 64 bit technology intel used was a system that flopped, they eventually just adopted for use all the AMD x86-64 instruction sets that AMD introduced (AMD was first btw). I cannot even recall what it was they called it...

I have a feeling this will go the same way...the players involved in HSA are too big and too many for Intel to overcome all of them. Much like Nvidia's CUDA is battling OpenCL, I think time will be on the foundation's side. There are many more adopters of OpenCL than there are for CUDA (scientific community is about all there is for CUDA adopters outside of individual companies).

I'm just going to say, its a little cocky to say that they were first. Yes they are, but trying to make that a point to try to win, is not right.

Also with CUDAand OpenCL, OpenCL is not as much as used as Dx11 and Dx10, which is what CUDA is better adapted for than OCL. AMD in a similar way is better with OpenCL than Dx. Food for Thought.

1.) It's not cocky...it's truth.

2.) OpenCL is not a competitor to DirectX in anyway shape or form...that's OpenGL.

3.) Nvidia and AMD both support DirectX11 and OpenGL 4.2.

4.) OpenCL is something completely different...and while NVidia supports this some, AMD is big on OpenCL. NVidia's counterpart is CUDA, which must be programmed for specifically to work on NVidia GPUs as it's proprietary. Any GPU can run OpenCL as it's an open standard...that's why CUDA will fail.

5.) OpenCL and CUDA are about using the GPU to do parallel processing by running compute functions on GPUs.

6.) DX and OpenGL are about 3D hardware acceleration and graphics APIs, as well as physics computations and other media tasks.

You might want to do some homework before you run around calling someone "cocky".

I have done my homework and understand what all this is. AMD does have the upper hand with OpenCL, and everyone knows that. However, how will CUDA fail if many corporations such as Bohemia Interactice, DICE, and many other are expanding with it? Most of the ArmA series is based off of Nvidia Physx which is greatly boosted with the more CUDA cores that you have.

Even so CUDA does still perform at a great level. The only advantage that AMD has is:
- obviously with video and photo editing
- cheaper, so its better for tighter budget people
- and games that use a majority of the OpenCL system

With all this, I may also add that with the architecture of games that are OpenCL based, isn't really catching up to any of the other programs. Yeah the game can sometimes run faster with it, but the quality to performance ratio is lacking on the quality side with it.

OpenCL will catch on more with each step of HSA integration. Hardware is not yet fully integrated...so how can OpenCL be widespread yet?

Additionally...CUDA is used by some scientific communities...however, notice that AMD has most of the design wins in the Supercomputers where GPGPU compute functions are most thoroughly utilized? You don't see many HPC's running Nvidia...there's a reason for that.

I agree totally with that. The power of AMD with what they are doing completely fits like a well made puzzle. However, on a gaming basis, its a completely different story, as you may know.

 

[GOM3RPLY3R]

GTX 670+3570K. LOLOLOLOLOLOLNO, more like 7750 or 6770 range.

I think he forgot where his bridge was...LOL.

PUT MY HEATSINK BACK ON!

You guys are funny. If you think that way, then I'll completely agree that AMD is a fashion company that shows more than it actually is. Thank you for this information, that may now be used against you. ^_^

 

[MU_Engineer]

I would target HD7770 performance, then its truely mainstream low end and 1080 ready. A while ago they said AMD would double the SP of Trinity that's a hair over 750 Stream processors for the top end part. current parts are 8 ROP's and 48 TMU's may need that and the memory bus bumped.

To double the SPs they'd have to jump to like 2Billion transistors (approx 125mil Transistors per CU). Couple that with the power draw of SR you'd have over a 200W CPU. Not likely without a hefty downclock of the CPU cores.

You're seeing why for PS4 they had to drop in the much smaller Jaguar cores to offset the 18 CU.

Adding a bunch of SPs to a current APU will not increase performance all that much because it is probably already bandwidth-limited. The APUs can use up to a 128 bit DDR3-2133 interface, good for 33.3 GB/sec. The HD 7750 with a 128 bit 1.125 GHz GDDR5 interface is good for 72 GB/sec. That's a pretty big difference in bandwidth. You'd need to add some L4/eDRAM or add additional memory channels to get much more performance out of the GPU part of the APU. That's also why the Jaguar setup in the PS4 has the supplemental GDDR5.

 

[juanrga]

I agree totally with that. The power of AMD with what they are doing completely fits like a well made puzzle. However, on a gaming basis, its a completely different story, as you may know.

AMD has the fastest GPU for gaming.

 

[mlscrow]

...Back to CPUs. I would like to talk about the new A10-6800k. I think this is isanely awesome. For something to take no more than 150 watts and have the power of an i5-3570k and a GTX 670 for about $150 is amazing.

^ex-AMD marketing analyst.

 

[8350rocks]

...Back to CPUs. I would like to talk about the new A10-6800k. I think this is isanely awesome. For something to take no more than 150 watts and have the power of an i5-3570k and a GTX 670 for about $150 is amazing.

^ex-AMD marketing analyst.

I think he's from a Fairy tale from the Grimm Brothers...the one about 3 billy goats and a bridge.

 

[griptwister]

http://www.quickmeme.com/meme/3uqowo/

Lol, and I just found this. I'm actually bummed because somewhere in the beginning this guy states that Kaveri is planned for next year... :/ So, I might buy a FX 8770 and a updated motherboard for the time being.

http://www.youtube.com/watch?feature=player_embedded&v=KE6pIMqMQ2E#!