AMD CPU speculation... and expert conjecture

[Tlight131]

why do people buy a cpu and upgrade it later?

Because two years from now technology will have changed and the want of the user pushes them to a new chip.

 

[-Fran-]

why do people buy a cpu and upgrade it later?

Initial cost is usually the answer. You have budget to a certain setup up *now*, and you know down the road will have money for upgrades (in particular, the CPU).

Problem is when you don't have an upgrade path and you think you do, I guess

In realistic terms, buying a s1155 Celeron/Pentium right now is a good move if you save money for a SB/IB CPU down the road. Either used or new. Intel already said they'll EOL IB in Q4-2013, so there's plenty of time for an upgrade. In the AMD camp, it's a mixed thing. AM3+ is EOL'ed already and you don't know if SR will be in AM3+ or be backwards with it, but you can still find low budget Athlon IIs and Phenom IIs that fit AM3+ and since PD-FX'es won't be EOLed soon, you can have a really cheap platform and an upgrade path until AMD decides what the hell it will do with SR (if they haven't already and I don't know about it xD). In the APU camp, AMD really played a bad card with FM1 and FM2. I know they had to make the jump and all, but it really screwed people (me) with FM1 sockets and non-upgradable HTPCs. FM2 should have a good life though, so buying something around it, should have life till SR's APU incarnation is born and stuff; that's Kaveri, right?

Cheers!

 

[de5_Roy]

for all amd users know, steamroller will be 28 nm. it will fit am3+ socket. it will have much higher ipc. it will be much better performing than bd/pd. global foundries will get the best yields for sr cpus in their history. it will bring salvation and save the mankind from jacking itself into the intel-matrix. it will be *gasp*.....the chosen one. 'cuz [strike]larry ellison[/strike] oracle has a profecy, or something. actually, no, he didn't. i made that part up. anyway, steamroller being the savior(that fits am3+ socket thus proves that amd has better upgradability (not sidegradability or downgradability) once and for all) is what's keeping the amd-faithful going. if steamroller doesn't fit am3+, they will be really upset.

@yuka: here's an old rumor link (keep the salt ready) http://semiaccurate.com/2012/11/19/amd-kills-off-big-cores-kaveri-steamroller-and-excavator/
kaveri should fit socket fm2. but amd is very tight lipped about sr recently. they essentially dropped the s-word from their p.r. and only mention kaveri.

 

[wh3resmycar]

Initial cost is usually the answer. You have budget to a certain setup up *now*, and you know down the road will have money for upgrades (in particular, the CPU).

Problem is when you don't have an upgrade path and you think you do, I guess

........

Cheers!

i would've get a used 2500k (i'd get a new psu with this though) or an i3-3220 to keep it conservative. those are my options, if only im not too lazy disassembling stuffs. on the AMD side though, when games are CPU intensive, whether you upgrade or not, you're at a disadvantage.

i'd probably take an AMD APU if only they'd take that 16x pci-e slot and cut the cost of a good board by half and reduce the price of the cpu itself. that way i won't be itching to stick a xx50/xx70 class GPU in there.

 

[cgner]

i would've get a used 2500k (i'd get a new psu with this though) or an i3-3220 to keep it conservative. those are my options, if only im not too lazy disassembling stuffs. on the AMD side though, when games are CPU intensive, whether you upgrade or not, you're at a disadvantage.

i'd probably take an AMD APU if only they'd take that 16x pci-e slot and cut the cost of a good board by half and reduce the price of the cpu itself. that way i won't be itching to stick a xx50/xx70 class GPU in there.

If it is a CPU intensive game that the developers actually bothered to code property and it uses all the cores u got to throw at it, like Battlefield 3, then AMD works just fine. Even older 1100T gets 60% load when running BF 3 on stock clocks and at 60 fps.

 

[sarinaide]

AMD wants socket stability at least until after the Steamroller arch, maybe revised chipsets and faster DIMMS and PCI-e will be bolted on but the socket 942 will still carry over at least until Steamroller.

 

[jdwii]

AMD wants socket stability at least until after the Steamroller arch, maybe revised chipsets and faster DIMMS and PCI-e will be bolted on but the socket 942 will still carry over at least until Steamroller.

Maybe after Steamroller Amd can i don't know put the north bridge on the CPU


I hope steamroller will be on AM3+ and my board since i can upgrade my 1100T.

 

[sarinaide]

I am dead sure I saw official AMD comments that 3rd generation parts were going to occupy the existing sockets.

 

[amdfangirl]

http://www.theinquirer.net/inquirer/news/2208525/amd-sticks-with-socket-am3-for-steamroller

 

[gamerk316]

If it is a CPU intensive game that the developers actually bothered to code property and it uses all the cores u got to throw at it, like Battlefield 3, then AMD works just fine. Even older 1100T gets 60% load when running BF 3 on stock clocks and at 60 fps.

But thats NEVER going to happen, because software isn't, and can't, be written that way. And I'm going to keep repeating this, because ITS TRUE.

Its the same old thing for almost half a decade now: E8600 or Q6600? i3-2300k or i5-2500k? For gaming, a fast dual will almost always be enough, with 64-player MP being the only scenario that really *needs* a quad.

Games are going to have two to three threads do 90% of the heavy lifting, simply because thats how it HAS to be done.

Pics via http://graphics.stanford.edu/~mdfisher/GPUView.html:

Main thread bounces between two CPU cores, and does 90% of the work, secondary thread mostly runs on one CPU, but does significantly less. Two threads do ALL the heavy lifting.

Same deal: Two main threads do all the work.

And I would imagine the same exact outcome in just about every game: 90% of the work being done by the main thread, with some parallel work by a secondary thread at the time of GPU rendering. Every other thread (out of a good 50) does negligible amounts of processing. But those two threads that actually control the program and handle the rendering? Thats ALWAYS going to be around, and theres nothing anyone can do about it.

But hey, feel free to ignore the software guy.

 

[esrever]

Maybe after Steamroller Amd can i don't know put the north bridge on the CPU
I hope steamroller will be on AM3+ and my board since i can upgrade my 1100T.

Im expecting they put northbridge onto the APU before trying it out on w/e follows AM3+. AMD probably isn't going to focus too much on the FX line in the future.

 

[noob2222]

Unfortunately until software programmers are taught multi-core programming from the start, you may be correct.

But if they start learning multi-core programming first, it will be easier for them to accept NEW WAYS of programming instead of being taught and stuck on the Von Newmann model.

One of the bigger problems is programmers are afraid to alienate their "dual core" fanbase.

The big mis-conception is having a program that can run on a dual core or having a program that CAN'T run on a dual core cpu. Until the latter happens, you are correct, but its not impossible as you say, its just not currently cost-effective to miss sales on all those dual core systems out there.

Times will change, they always do.

 

[JAYDEEJOHN]

One word
Game engines

 

[JAYDEEJOHN]

As for dual cores, and 2 threads
http://forum.beyond3d.com/showpost.php?p=1702005&postcount=10

 

[gamerk316]

Unfortunately until software programmers are taught multi-core programming from the start, you may be correct.

But if they start learning multi-core programming first, it will be easier for them to accept NEW WAYS of programming instead of being taught and stuck on the Von Newmann model.

One of the bigger problems is programmers are afraid to alienate their "dual core" fanbase.

The big mis-conception is having a program that can run on a dual core or having a program that CAN'T run on a dual core cpu. Until the latter happens, you are correct, but its not impossible as you say, its just not currently cost-effective to miss sales on all those dual core systems out there.

Times will change, they always do.

Please noob, STOP IT. I've explained dozens of times now the reasons you will not see most program loads scale beyond more then a handful of cores, the simplest being, that due to the way the hardware, OS, API's, and DirectX work, that its impossible to end up with more then 2-3 threads that do the overwhelming majority of the work. You are going to have your main control thread and rendering thread as the big two. Everything else (audio, AI, UI, and most Physics implementations) are non-factors computationally; they take so few resources you can generally assume they are not impacting the results any.

Secondly, as I've pointed out several times already: If no CPU core is loaded to 100%, guess what? Adding more cores isnt' going to make the program in question run any faster; you aren't CPU bottlenecked. Thats how the i3 still competes: Its fast enough, even with two cores, for 90% of all games, to get all is work done.

Thirdly, you do not "teach" multi-core programming; core loading is left up to the OS scheduler, and its under VERY carefully controlled circumstances where you can tell the OS that you know more about its internals then the people who designed it. Most devs just leave the default OS mask (all cores enabled) and let the OS schedule away. Only in the case where extensive testing has shown two threads to be both high workload and totally parallel should efforts be taken to ensure they always operate on separate cores (and that is not as easy as you think; heaven forbid your AV decides that exact moment to go off, just as you locked one of the two threads to the same core the AV is using.)

Fourthly, due to various bottlenecks on the OS side of the house, its a well known fact that overthreading, even for reasonably parallel applications, will eventually start to reduce performance. So doing more threading, without proper benchmarking and testing, and validating an actual benefit, is a waste of time, resources, money, and program stability (in which case, you'll complain about how lazy devs were to release a game in a non-working state).


Now, if you have some magical framework that gets around these "minor" problems, feel free to post them. I for one, as a software developer, am REALLY starting to get annoyed by people who have probably never written a SLOC for an actual program in their life telling me that years of experience working with the internals of Windows is (among other OS's) is all incorrect.

 

[JAYDEEJOHN]

gamers right
There needs to be fundamental changes, similar to what we may see in the new consoles/HSA etc, and even there, porting doesnt change this really

 

[JAYDEEJOHN]

http://www.youtube.com/watch?v=wCaFMSfYLDA&list=PLxlXCOZVBRMIAcv9QnzfmQU4e-yQIDTEj

 

[sonoran]

Unfortunately until software programmers are taught multi-core programming from the start, you may be correct.

But if they start learning multi-core programming first, it will be easier for them to accept NEW WAYS of programming instead of being taught and stuck on the Von Newmann model.

So what non Von-Neumann computer architecture and software programming paradigm do you suggest they teach?

 

[viridiancrystal]

Please noob, STOP IT. I've explained dozens of times now the reasons you will not see most program loads scale beyond more then a handful of cores, the simplest being, that due to the way the hardware, OS, API's, and DirectX work, that its impossible to end up with more then 2-3 threads that do the overwhelming majority of the work. You are going to have your main control thread and rendering thread as the big two. Everything else (audio, AI, UI, and most Physics implementations) are non-factors computationally; they take so few resources you can generally assume they are not impacting the results any.

Secondly, as I've pointed out several times already: If no CPU core is loaded to 100%, guess what? Adding more cores isnt' going to make the program in question run any faster; you aren't CPU bottlenecked. Thats how the i3 still competes: Its fast enough, even with two cores, for 90% of all games, to get all is work done.

Thirdly, you do not "teach" multi-core programming; core loading is left up to the OS scheduler, and its under VERY carefully controlled circumstances where you can tell the OS that you know more about its internals then the people who designed it. Most devs just leave the default OS mask (all cores enabled) and let the OS schedule away. Only in the case where extensive testing has shown two threads to be both high workload and totally parallel should efforts be taken to ensure they always operate on separate cores (and that is not as easy as you think; heaven forbid your AV decides that exact moment to go off, just as you locked one of the two threads to the same core the AV is using.)

Fourthly, due to various bottlenecks on the OS side of the house, its a well known fact that overthreading, even for reasonably parallel applications, will eventually start to reduce performance. So doing more threading, without proper benchmarking and testing, and validating an actual benefit, is a waste of time, resources, money, and program stability (in which case, you'll complain about how lazy devs were to release a game in a non-working state).


Now, if you have some magical framework that gets around these "minor" problems, feel free to post them. I for one, as a software developer, am REALLY starting to get annoyed by people who have probably never written a SLOC for an actual program in their life telling me that years of experience working with the internals of Windows is (among other OS's) is all incorrect.

But thats NEVER going to happen, because software isn't, and can't, be written that way. And I'm going to keep repeating this, because ITS TRUE.

Its the same old thing for almost half a decade now: E8600 or Q6600? i3-2300k or i5-2500k? For gaming, a fast dual will almost always be enough, with 64-player MP being the only scenario that really *needs* a quad.

Games are going to have two to three threads do 90% of the heavy lifting, simply because thats how it HAS to be done.

Pics via http://graphics.stanford.edu/~mdfisher/GPUView.html:

http://graphics.stanford.edu/~mdfisher/Images/GPUViewLargeRenderTime.png

Main thread bounces between two CPU cores, and does 90% of the work, secondary thread mostly runs on one CPU, but does significantly less. Two threads do ALL the heavy lifting.

http://graphics.stanford.edu/~mdfisher/Images/GPUViewWow.png

Same deal: Two main threads do all the work.

And I would imagine the same exact outcome in just about every game: 90% of the work being done by the main thread, with some parallel work by a secondary thread at the time of GPU rendering. Every other thread (out of a good 50) does negligible amounts of processing. But those two threads that actually control the program and handle the rendering? Thats ALWAYS going to be around, and theres nothing anyone can do about it.

But hey, feel free to ignore the software guy.

Well, 90% of all games includes any that were made before multi-core processing, which means you're most likely wrong, it is probably much more than 90%.

Also, I still want you to answer my question: If parallelism is so difficult, why did we ever leave single core processors? Intel, AMD, Samsung, Nvidia, Qualcomm, and others, they all make multi-core processors. If software scaling is so impossible to achieve, why do they do this?

 

[cgner]

But thats NEVER going to happen, because software isn't, and can't, be written that way. And I'm going to keep repeating this, because ITS TRUE.

Its the same old thing for almost half a decade now: E8600 or Q6600? i3-2300k or i5-2500k? For gaming, a fast dual will almost always be enough, with 64-player MP being the only scenario that really *needs* a quad.

Games are going to have two to three threads do 90% of the heavy lifting, simply because thats how it HAS to be done.

Pics via http://graphics.stanford.edu/~mdfisher/GPUView.html:

http://graphics.stanford.edu/~mdfisher/Images/GPUViewLargeRenderTime.png

Main thread bounces between two CPU cores, and does 90% of the work, secondary thread mostly runs on one CPU, but does significantly less. Two threads do ALL the heavy lifting.

http://graphics.stanford.edu/~mdfisher/Images/GPUViewWow.png

Same deal: Two main threads do all the work.

And I would imagine the same exact outcome in just about every game: 90% of the work being done by the main thread, with some parallel work by a secondary thread at the time of GPU rendering. Every other thread (out of a good 50) does negligible amounts of processing. But those two threads that actually control the program and handle the rendering? Thats ALWAYS going to be around, and theres nothing anyone can do about it.

But hey, feel free to ignore the software guy.

lol ignore u? We say the same thing every time, and I repeated it for the other guy Also if 2300k was available, I would hands down pick it over any other high end Intel quad core chips.

 

[jdwii]

Well, 90% of all games includes any that were made before multi-core processing, which means you're most likely wrong, it is probably much more than 90%.

Also, I still want you to answer my question: If parallelism is so difficult, why did we ever leave single core processors? Intel, AMD, Samsung, Nvidia, Qualcomm, and others, they all make multi-core processors. If software scaling is so impossible to achieve, why do they do this?

Gamer i guess this guy has never heard of multitasking or marketing.

 

[jdwii]

Im expecting they put northbridge onto the APU before trying it out on w/e follows AM3+. AMD probably isn't going to focus too much on the FX line in the future.

Pretty sure the APU already has a northbridge on die. And i hope we don't come to that point but it seriously seems that way.

 

[jdwii]

Also in a bigger note and more importantly

If Amd new this and we all know they did they can't be retarded being CPU engineers. Then why would they make the module design instead of a stronger cores? Instead of blaming programmers and telling them to do the impossible maybe we should blame Amd for making a CPU the way they did.


No i'm sure Opencl or HSA computing is the next step. We will still have 2-3 cores for gaming but some things will be put on the GPU instead such as physics. Maybe with the next consoles will get some change on this since i think even the Wii U can do it.

 

[noob2222]

Please noob, STOP IT..

Explain battlefield 3.
explain Metro 2033
explain Civ V.

those are all very much multi-core friendly. All you ever do is say "oh, they run on a dual core system, so they are only a dual core game and it will NEVER happen."

you want to cry foul, fine. explain how these games are multi-core friendly today when it will NEVER happen.

this is obviously fake since no game uses more than 2 cores, because the dual core system is unplayable. Single player benchmarks has shown that its perfect in every way.

If you want me to stop it, explain how games will NEVER be multi-core friendly when they are that way TODAY.

Kinda funny to me because it seems your on a crusade to keep programming simple, going against the way that hardware is headed. From tsx to lock-free to parallel. nothing is acceptable because it would require re-learning what you alredy know.

Fight the future, fight innovation, fight advancements in programming, Fight to protect the dual core systems.

 

[skitz9417]

http://www.tomshardware.com/news/AMD-Steamroller-Piledriver-Kaveri-processors,17217.html