AMD CPU speculation... and expert conjecture

Page 57 - Seeking answers? Join the Tom's Hardware community: where nearly two million members share solutions and discuss the latest tech.
Status
Not open for further replies.

jdwii

Splendid



Yes to a point but look at the charts 9/10 Amd and Intel are even(which was not your point)

A A10 and a 8350 are pretty even and when they aren't they both give playable FPS.

Now look at the I3 and the I7 its even closer.




One CPU was introduced in 2004 and the other in 2006. At that note i guess the 9800gt was a rip off when the 6850 came out 2 years later?

Also when Intel and Amd where going at it Amd still had better price/performance then intel. Heck i'll take a sempron from 2003 over a celeron in 2003 any day.



I think software engineers know more about programming then engineers. I'm sorry to say this but their is a reason why they went with a module design vs a more efficient one they could not compete with Intel head on with IPC. If gamer was wrong then why just why has it been over 7 years and most games are still dual threaded why is a I3 almost always next to a I7 why does skyrim only use 2 cores?

The only thing about having a CPU over 4 cores useful is if you game and do other things in the background or you do other CPU heavy tasks.
 
FarCry3.png

look at them dual cores rock farcry 3. Especially the ones without hyper threading.
 

Skyrim's engine is still based on the oblivion engine which is old they didn't really even try to make the game run well. There is no money in optimizing for more cores and denying a large part of potential buyers for the past 7 years. 4 cores are becoming the norm and now we are finally getting some games that use them. Most of the latest games built on DX11 are being bottlenecked by 2 threads. Farcry 3, hitman absolution, dirt showdown, battlefield 3, max payne 3, ect. Its slow but moving. Unless you are telling me that the cryengine is doing the impossible on farcry 3 and using more than 2 hardware threads. By Gamer's logic, the newer games using more cores is impossible and that the pentium dual cores should play the games better than the FX8350 due to the better IPC.
 

jdwii

Splendid



The thing is though the I3 is usually even with the 8350 if not better.
 

kettu

Distinguished
May 28, 2009
243
0
18,710


Well, you said "Funny, I have't seen a single example of that as far as games go..." BF3 multiplayer is a game is it not? So there's your example.

As far as your speculation about clockspeeds... Why bother with that. Adding parallel execution resources seem to do the trick pretty much proving that this particular game can effectively utilise atleast six cores.

About the i3:
Why would I introduce another variable in the mix (a completely different architecture) when adding parallel (that's the magick word here) resources even when lowering single threaded performance (via lower clockspeed of x6) increases performance.
 

i3 runs 4 threads. Game is only optimized for 4 threads. Better IPC for the i3. There is balance. Why are you using the i3 for comparison and not a pentium if you are arguing 2 hardware threads is all you need?
 

earl45

Distinguished
Nov 10, 2009
434
0
18,780
 

Cazalan

Distinguished
Sep 4, 2011
2,672
0
20,810



The game itself uses 2 hardware threads but the OS still needs some CPU power. That's where even the Athlon X3 kicks in a sizable 50% bump over the Trinity dual cores.

The saddest thing about that graph is just how bad a single module Trinity is.

AMD really needs those Steamroller improvements. It's a shame it's been delayed so long.
 
I think software engineers know more about programming then engineers. I'm sorry to say this but their is a reason why they went with a module design vs a more efficient one they could not compete with Intel head on with IPC. If gamer was wrong then why just why has it been over 7 years and most games are still dual threaded why is a I3 almost always next to a I7 why does skyrim only use 2 cores?

It's about the evolving process's behind development. It often takes years to develop a "game" and their going to be targeting the 95% market share not the 5%. This is why quad cores are now the standard where as years ago dualies were the standard and before that single core CPUs. Six / Eight cores start becoming the standard in the next five or so years. I mentioned above what it takes to develop a program to run scale well on parallel architectures, you need to redefine your problem in a way that can be broken into multiple simultaneous tasks. There is only so much you can do if the original concept and algorithm was designed to run in serial.

It's the sad truth that hardware has developed far faster then the software design methods behind it. Also be very careful with i3 benchmarks of a single player game running in a timed demo mode. If an i5 isn't pulling 200% of the performance numbers of an i3 then whats happening is the i3 is hitting 100% while the i5 is sitting there with underutilized processor power and therefor isn't indicative of actual performance capacity.
 
AMD really needs those Steamroller improvements. It's a shame it's been delayed so long.

I'm glad that Richland is coming out soon, that will tell us what improvements have been made. APU's are the future for them and most of the market. You need to put more capability into an ever smaller package.
 

Cazalan

Distinguished
Sep 4, 2011
2,672
0
20,810



Unfortunately we won't see Steamroller cores until Q4-2013 with Kaveri.

Richland may have some surprises in store but it's just a tuned Trinity, not the major changes that Steamroller has.

Funny looking back how close Charlie was in August.

http://semiaccurate.com/2012/08/06/amd-to-update-trinity-to-trinity-2-0/



 


Yeah for some reason I was thinking Richland was an architecture change. Looks to just be some fine tuning though if they ever manage to get the L2 latency down it'll be a noticeable change.
 

jdwii

Splendid



Since i find HT to be irrelevant for those games. And even if you a Pentium was rated better then other cheaper gaming CPU's on toms.

Age has nothing to do with my point at all, at the time you couldn't buy anything better from AMD then the FX-52 and it was 1000.00 dollars, Intel came out with E6600 which is a better performing chip for 700.00 less then the FX-52.
So my point was Intel gave all of us the cheap CPU's we enjoy today.

I don't think you understand that technology gets cheaper to produce with age(Intel also had better fabs and could easily make that for cheap)

And who even bought that processor? Their was slightly cheaper Athlons that were almost(90% as good).

The Athlons where better then the Intel Pentiums anyways and that is where Intel fought the Price/Performance.
 

jdwii

Splendid



I'm hoping the reason for richlands is to improve on steamroller.

 

mayankleoboy1

Distinguished
Aug 11, 2010
2,497
0
19,810
If a thread is being scheduled to run on a core, but is locked out on another thread, will that make the OS count that core as 100% utilized ? If so, that can account for ssome games using 4 cores in task manager.
 

earl45

Distinguished
Nov 10, 2009
434
0
18,780

If this is your argument the FX3850 would cost more then any i7's but it don't, why is that?
The bottom line is you can't charge a premium for a chip that the competitions medium chip
can best it for less.
 

Cazalan

Distinguished
Sep 4, 2011
2,672
0
20,810



The evidence points to a lack of high performance 28nm process being ready.

Neither TSMC nor GF are shipping 28nm chips running past 3Ghz.

AMDs first 28nm APU will be Kabini running at roughly 2Ghz.
 


Depends on the OS.

Windows NT schedule is psychotic IMO.

OS's use something known as preemptive multitasking to simulate mult-tasking. A single "register stack" aka "core" can only process one instruction stream at a time. What the OS does is create predetermined time slices and each thread is then assigned a "core" to run it's instructions on during that time slice. When it's scheduled time has been exhausted the control is returned to the OS which then schedules another instruction stream to run. This happens extremely fast, so fast that user's will never notice the OS switching between hundred's of threads. Processor "utilization" is a measurement of how many of those time slices are being used to run code and how many are going underutilized. Many programs have wait states and other bottle necks implemented due to I/O needs. A program makes a request to the OS to open a file and read some data then goes idle while all that needs to be done. During that time there will be periods where nothing is happening (takes time to spin the disk and for the head to seek), during that time it'll process the music your playing.

Now on an underutilized system there is a ton of free processing resources and that's why you see the usage so low. On a system that's under load with multiple active threads asking for processor attention the OS tries to evenly assign time. The Windows NT scheduler use's priority as a weight with "real time" being "give 100% to this process with no breaks". Two programs with the same weight will be given the same amount of processor time under load. If your load is larger then the amount of resources you have you'll start to see stuttering and lagging of the UI. Back in the single core days it was possible for extreme load to lock the UI by minimizing the amount of time the UI got to run on the CPU.

This should explain why having a moderate amount of cores is ideal. It also explains why applications that have all their heavy code in a single thread don't scale well. They can utilize at most 100% of a single "core". thus on a four core system that application will never use more then 25% of that CPU's power. Also HT does not create "virtual cores" or other nonsense. All it does is create two register stacks per "core", the OS identifies and handles each register stack as an individual CPU and assigns threads to it accordingly. Both of those register stacks run on the exact same hardware, the internal CPU scheduler will divvy up the processor resources between the two. As a single thread will rarely utilize all 100% of a "cores" available resources HT becomes a method to more efficiently utilize those resources.


This is important to keep in mine when your looking at the newer AMD architecture. You can have an eight "core" CPU (16 ALU's worth of integer performance) but if an application has all it's heavy lifting on one to two threads then your only effectively utilizing 12~25% of that CPU's power while the rest is just sitting there doing minimal work. It also explains why the Intel SB / IB four core CPUs (12 ALU's worth of integer performance) are able to beat the AMD eight core CPUs. The Intel ones are simply more efficient at maximizing the utilization of processor resources.
 

truegenius

Distinguished
BANNED


gamerk316 wrote:
1: I explicity states that Multiplayer lends itself to better performance with more cores, since you are adding a fairly process heavy thread into the mix.

got my point ;)
i mean, gamerk316 already answered it

and

esrever wrote :

i3 runs 4 threads. Game is only optimized for 4 threads. Better IPC for the i3. There is balance. Why are you using the i3 for comparison and not a pentium if you are arguing 2 hardware threads is all you need?
gamerk316 wrote:

I'm not saying quads don't help to some extent, but
for the majority of games, an i3 is more then
sufficient. In most cases, you run into a brick wall
in terms of scaling after 3 cores or so.

got my point ;)



i mean, read previous posts carefully, you may found your answers there

:D glad i am not gamerk316 , otherwise i will be dead because of this much repeating of same thing again and again and again :p
 
Running into a brick wall above 3 cores means the athlon x3s are only getting 70% of the FSP as the quad cores on farcry 3. The threading will be there one day to use 8 core, then 16 then 32 cores then maybe 64 cores, above that Gamerk would have a point because amdahl's law actually becomes significant as programs with below 80% parallel code would have very poor gain. Currently the only reason we don't have many core use is because of the way most programers approach a program. Locks and stalls are hard to deal with but getting these things down is not impossible.

How are developers even developing on a ps3 if all they could do is use 3 cores? They work around the hardware and with enough funding, they actually can make a game with it. PC doesn't get the funding and focus of developers and games don't use all available resources, it doesn't mean using the resources is impossible.
 
Status
Not open for further replies.