Overclock.net › Forums › Graphics Cards › NVIDIA › MSI GTX 560ti TwinFrozr II: Phenom x4 945 vs. i5-2500k
New Posts  All Forums:Forum Nav:

MSI GTX 560ti TwinFrozr II: Phenom x4 945 vs. i5-2500k - Page 2

post #11 of 23
I wouldn't recommend a platform change for a single 560 before, and after seeing these tests I wouldn't in the future either.

You can see my quote in Heavy MG sig, you need to balance your system.

Now that you have a i5 however, you can upgrade to more gpu power and see a better performance difference between the two platforms.
    
CPUMotherboardGraphicsGraphics
Intel Core i5 2500K P8P67 PRO NVIDIA GeForce GTX 470 NVIDIA GeForce GTX 470 
GraphicsRAMRAMRAM
NVIDIA GeForce 9800 GT G-Skill A-Data G-Skill 
RAMHard DriveOptical DriveOS
A-Data Crucial M4 64GB + 1TB F3 Spinpoint $155 LS/DL DVD RW $?? Windows 8 64-bit "Epic Registry" Edition 
MonitorPowerCase
ASUS 21.5 1920x1080 2ms $135 CORSAIR HX850 $120 Mother Earth $free 
  hide details  
Reply
    
CPUMotherboardGraphicsGraphics
Intel Core i5 2500K P8P67 PRO NVIDIA GeForce GTX 470 NVIDIA GeForce GTX 470 
GraphicsRAMRAMRAM
NVIDIA GeForce 9800 GT G-Skill A-Data G-Skill 
RAMHard DriveOptical DriveOS
A-Data Crucial M4 64GB + 1TB F3 Spinpoint $155 LS/DL DVD RW $?? Windows 8 64-bit "Epic Registry" Edition 
MonitorPowerCase
ASUS 21.5 1920x1080 2ms $135 CORSAIR HX850 $120 Mother Earth $free 
  hide details  
Reply
post #12 of 23
Thread Starter 
Quote:
Originally Posted by LOL_Wut_Axel View Post
That's 3 FPS. But anyway, very nice system. 2500K goes very well with a 560 Ti.
hahaha, yea i just noticed that. i guess my mathematical skills were impaired after a couple beers last night.

im going to go back and check the actual bench scores. i dont know where the hell i came up with 8.
My System
(13 items)
 
  
CPUMotherboardGraphicsRAM
2500k - 4.8ghz @1.35v p67a-g45 MSI gtx560ti 2x2 patriot viper 1600mhz 
Hard DriveOSMonitorPower
3 WD's - 1.7TB Total Win7 Ultimate ViewSonic 2702@1920x1080 & Mitsu 65" @1920x1080 BFG ls550 
Case
800D Obsidian 
  hide details  
Reply
My System
(13 items)
 
  
CPUMotherboardGraphicsRAM
2500k - 4.8ghz @1.35v p67a-g45 MSI gtx560ti 2x2 patriot viper 1600mhz 
Hard DriveOSMonitorPower
3 WD's - 1.7TB Total Win7 Ultimate ViewSonic 2702@1920x1080 & Mitsu 65" @1920x1080 BFG ls550 
Case
800D Obsidian 
  hide details  
Reply
post #13 of 23
Thread Starter 
Quote:
Originally Posted by CULLEN View Post
$135 CPU @ 3.7 GHz vs $225 CPU @ 4.5 GHz

I think I know who's going to win this, however no need for the 2500K to be so high since it bottle necking GTX560 hard.
true, and technology does change. but i bought that phenom within the last year and it retailed for over $200 when it came out.


i know the gpu is a bottleneck, but i do video/audio converting and stuff like that. its not overclocked just for games.
Edited by stRodda - 4/19/11 at 7:10pm
My System
(13 items)
 
  
CPUMotherboardGraphicsRAM
2500k - 4.8ghz @1.35v p67a-g45 MSI gtx560ti 2x2 patriot viper 1600mhz 
Hard DriveOSMonitorPower
3 WD's - 1.7TB Total Win7 Ultimate ViewSonic 2702@1920x1080 & Mitsu 65" @1920x1080 BFG ls550 
Case
800D Obsidian 
  hide details  
Reply
My System
(13 items)
 
  
CPUMotherboardGraphicsRAM
2500k - 4.8ghz @1.35v p67a-g45 MSI gtx560ti 2x2 patriot viper 1600mhz 
Hard DriveOSMonitorPower
3 WD's - 1.7TB Total Win7 Ultimate ViewSonic 2702@1920x1080 & Mitsu 65" @1920x1080 BFG ls550 
Case
800D Obsidian 
  hide details  
Reply
post #14 of 23
Thread Starter 
Quote:
Originally Posted by BallaTheFeared View Post
I wouldn't recommend a platform change for a single 560 before, and after seeing these tests I wouldn't in the future either.

You can see my quote in Heavy MG sig, you need to balance your system.

Now that you have a i5 however, you can upgrade to more gpu power and see a better performance difference between the two platforms.
true true. im sure the future holds a second 560 and of course a new psu.

normally i wouldnt have. but i had a sort of hookup on the cpu(not stolen) and store credit at frys electronics, so i put $90 towards the mobo. so in the end, i didnt pay that much for a change over. and i'll be selling the amd board/cpu/cooler, so i might actually make money on the whole transaction.
My System
(13 items)
 
  
CPUMotherboardGraphicsRAM
2500k - 4.8ghz @1.35v p67a-g45 MSI gtx560ti 2x2 patriot viper 1600mhz 
Hard DriveOSMonitorPower
3 WD's - 1.7TB Total Win7 Ultimate ViewSonic 2702@1920x1080 & Mitsu 65" @1920x1080 BFG ls550 
Case
800D Obsidian 
  hide details  
Reply
My System
(13 items)
 
  
CPUMotherboardGraphicsRAM
2500k - 4.8ghz @1.35v p67a-g45 MSI gtx560ti 2x2 patriot viper 1600mhz 
Hard DriveOSMonitorPower
3 WD's - 1.7TB Total Win7 Ultimate ViewSonic 2702@1920x1080 & Mitsu 65" @1920x1080 BFG ls550 
Case
800D Obsidian 
  hide details  
Reply
post #15 of 23
Of course it's going to depend on the testing scenario: the degree of CPU-dependency (meaning 'how much work does the GPU require of the CPU per frame rendered') the test has inherently, along with what IQ settings (res, AA, etc) you run on that test (of course the higher the settings, the lower the FPS, and hence the less likely you are to have frames BN'ed by the CPU performance).

3dMark06 is so old that with modern cards it runs at FPS far in excess of what the cards it was designed to test were capable of ... so the GPU tests have become CPU bottlenecked (esp. tests one and two).

And as you can see Vantage, at least in Performance mode, is headed in the same direction. However, if you test in High mode, or better yet Extreme (which are much more akin to settings you'd actually run w/a GPU like the 560) you won't see much difference between the two cpus.

Some of the other games that are likely to show measurable difference in Perf between them are: Crysis, BFBC2, Black Ops, and of course GTAIV.

General-speaking though any decent, well OC'd quad is quite adequate for any single-card setup.

Lastly ... just looking at averages doesn't really tell the whole story, because frequently the power of the CPU will dictate the MIN fps encountered during the test. So ... a loss of an average 3 fps over the course of a test may not seem like much (and it may indeed not mean much), but there exists a possibility that it actually IS important.

Imagine if 95% of the test ran exactly the same, but because of the slower cpu, there was a 5 second stretch that dropped to 10fps, whereas the better CPU only dropped to 25fps.

In this case, the 3fps difference in average is actually much more important than it seems, as we're talking about going from a totally 'playable' scenario to one where if you played the game w/the slower CPU, every now and again, you're going to encounter stretches that are totally laggy. And these could well come at 'bad times'.

So bottom-line when comparing CPU's you really want to look at MIN's as well as average FPS.
    
CPUMotherboardGraphicsRAM
xeon X5675 6-core @ 4.1ghz (1.29v, 20x205 +ht ) rampage iii extreme msi rx470 gaming X (the $159 budget king) 3 x 2gb corsair xms3 pc12800 (9-9-9-24-1T@1600MHz) 
Hard DriveOptical DriveCoolingOS
hynix 250gb ssd (boot), 2tb deskstar (apps),1tb... plextor px-712sa - still the best optical drive... corsair h8o v2 aio W10 home 
MonitorPowerCaseAudio
asus vw266h 25.5" (1920x1200) abs sl (enermax revolution) * single 70A rail 850w silverstone rv-03 XFi Titanium 
  hide details  
Reply
    
CPUMotherboardGraphicsRAM
xeon X5675 6-core @ 4.1ghz (1.29v, 20x205 +ht ) rampage iii extreme msi rx470 gaming X (the $159 budget king) 3 x 2gb corsair xms3 pc12800 (9-9-9-24-1T@1600MHz) 
Hard DriveOptical DriveCoolingOS
hynix 250gb ssd (boot), 2tb deskstar (apps),1tb... plextor px-712sa - still the best optical drive... corsair h8o v2 aio W10 home 
MonitorPowerCaseAudio
asus vw266h 25.5" (1920x1200) abs sl (enermax revolution) * single 70A rail 850w silverstone rv-03 XFi Titanium 
  hide details  
Reply
post #16 of 23
Thread Starter 
^ such a long explanation, I'm not going to quote it, but thanks for that. It does make complete sense. You always want a nice high steady average, but if you're dipping down every once in a while, which will always be at key moments, since that's when all the action is happening onscreen, it'll suck.


Are there any programs that will plot a graph of the current fps? I'm not sure why, but the fps in afterburner doesn't work anymore, it just stays blank.
My System
(13 items)
 
  
CPUMotherboardGraphicsRAM
2500k - 4.8ghz @1.35v p67a-g45 MSI gtx560ti 2x2 patriot viper 1600mhz 
Hard DriveOSMonitorPower
3 WD's - 1.7TB Total Win7 Ultimate ViewSonic 2702@1920x1080 & Mitsu 65" @1920x1080 BFG ls550 
Case
800D Obsidian 
  hide details  
Reply
My System
(13 items)
 
  
CPUMotherboardGraphicsRAM
2500k - 4.8ghz @1.35v p67a-g45 MSI gtx560ti 2x2 patriot viper 1600mhz 
Hard DriveOSMonitorPower
3 WD's - 1.7TB Total Win7 Ultimate ViewSonic 2702@1920x1080 & Mitsu 65" @1920x1080 BFG ls550 
Case
800D Obsidian 
  hide details  
Reply
post #17 of 23
Quote:
Originally Posted by brettjv View Post
Lastly ... just looking at averages doesn't really tell the whole story, because frequently the power of the CPU will dictate the MIN fps encountered during the test. So ... a loss of an average 3 fps over the course of a test may not seem like much (and it may indeed not mean much), but there exists a possibility that it actually IS important.

Imagine if 95% of the test ran exactly the same, but because of the slower cpu, there was a 5 second stretch that dropped to 10fps, whereas the better CPU only dropped to 25fps.

In this case, the 3fps difference in average is actually much more important than it seems, as we're talking about going from a totally 'playable' scenario to one where if you played the game w/the slower CPU, every now and again, you're going to encounter stretches that are totally laggy. And these could well come at 'bad times'.

So bottom-line when comparing CPU's you really want to look at MIN's as well as average FPS.

Actually that is exactly why I left AMD for my i5-2500k. While my average wasn't all that far behind with a single 470 in many benchmarks I saw some huge bottlenecks - for instance in vantage in the first tests it would go up and down all over the place very deeply to the point of 60% gpu bottlenecks, and in the second test it was smoother but at the part where the camera is following the ship my gpu usage would slowly drop below 70%. While it wouldn't have been so bad if those bottlenecks occurred around my highest fps, they weren't. They were happening when fps were going their lowest.

I would have been fine in most games but when I started to get the itch for SLI and both under water I knew my 1090T was going to have serious issues in many of the games I play.


To track gpu usage and fps on a graph use MSI Afterburner.
    
CPUMotherboardGraphicsGraphics
Intel Core i5 2500K P8P67 PRO NVIDIA GeForce GTX 470 NVIDIA GeForce GTX 470 
GraphicsRAMRAMRAM
NVIDIA GeForce 9800 GT G-Skill A-Data G-Skill 
RAMHard DriveOptical DriveOS
A-Data Crucial M4 64GB + 1TB F3 Spinpoint $155 LS/DL DVD RW $?? Windows 8 64-bit "Epic Registry" Edition 
MonitorPowerCase
ASUS 21.5 1920x1080 2ms $135 CORSAIR HX850 $120 Mother Earth $free 
  hide details  
Reply
    
CPUMotherboardGraphicsGraphics
Intel Core i5 2500K P8P67 PRO NVIDIA GeForce GTX 470 NVIDIA GeForce GTX 470 
GraphicsRAMRAMRAM
NVIDIA GeForce 9800 GT G-Skill A-Data G-Skill 
RAMHard DriveOptical DriveOS
A-Data Crucial M4 64GB + 1TB F3 Spinpoint $155 LS/DL DVD RW $?? Windows 8 64-bit "Epic Registry" Edition 
MonitorPowerCase
ASUS 21.5 1920x1080 2ms $135 CORSAIR HX850 $120 Mother Earth $free 
  hide details  
Reply
post #18 of 23
Thread Starter 
Quote:
Originally Posted by BallaTheFeared View Post
Actually that is exactly why I left AMD for my i5-2500k. While my average wasn't all that far behind with a single 470 in many benchmarks I saw some huge bottlenecks - for instance in vantage in the first tests it would go up and down all over the place very deeply to the point of 60% gpu bottlenecks, and in the second test it was smoother but at the part where the camera is following the ship my gpu usage would slowly drop below 70%. While it wouldn't have been so bad if those bottlenecks occurred around my highest fps, they weren't. They were happening when fps were going their lowest.

I would have been fine in most games but when I started to get the itch for SLI and both under water I knew my 1090T was going to have serious issues in many of the games I play.


To track gpu usage and fps on a graph use MSI Afterburner.
I had just mentioned above, but the fps graph in afterburner doesn't work for some reason.

And about gpu usage, if it shows 100% usage, does that mean there isn't a bottleneck? Or is it possible that the gpu is putting out 100%, but that a better CPU will raise fps by making other areas of the system run more efficiently?
My System
(13 items)
 
  
CPUMotherboardGraphicsRAM
2500k - 4.8ghz @1.35v p67a-g45 MSI gtx560ti 2x2 patriot viper 1600mhz 
Hard DriveOSMonitorPower
3 WD's - 1.7TB Total Win7 Ultimate ViewSonic 2702@1920x1080 & Mitsu 65" @1920x1080 BFG ls550 
Case
800D Obsidian 
  hide details  
Reply
My System
(13 items)
 
  
CPUMotherboardGraphicsRAM
2500k - 4.8ghz @1.35v p67a-g45 MSI gtx560ti 2x2 patriot viper 1600mhz 
Hard DriveOSMonitorPower
3 WD's - 1.7TB Total Win7 Ultimate ViewSonic 2702@1920x1080 & Mitsu 65" @1920x1080 BFG ls550 
Case
800D Obsidian 
  hide details  
Reply
post #19 of 23
If it shows 99% usage than the bottleneck is the gpu, which is good.

A better cpu will keep it at 99% more often, or always in your case I don't think you'll ever see it drop much below that. Having more cpu power than your gpu can handle just means you've opened the doors to greater performance and amazing scaling in SLI.

Some AMD people mock me because I said it, but I honestly feel as a gamer you only need as much cpu power as your gpu(s) can handle. If your SB chip can keep your 560 at 99% all the time at 4GHz than overclocking it to 5GHz isn't going to do anything for you as far as fps go.

There are sometimes other reasons that cause low gpu usage, but generally if others do not have the same problem with the same hardware but have better cpus or higher overclocks you can assume the lower gpu usage is because the cpu can't push it enough.
Edited by BallaTheFeared - 4/19/11 at 11:38pm
    
CPUMotherboardGraphicsGraphics
Intel Core i5 2500K P8P67 PRO NVIDIA GeForce GTX 470 NVIDIA GeForce GTX 470 
GraphicsRAMRAMRAM
NVIDIA GeForce 9800 GT G-Skill A-Data G-Skill 
RAMHard DriveOptical DriveOS
A-Data Crucial M4 64GB + 1TB F3 Spinpoint $155 LS/DL DVD RW $?? Windows 8 64-bit "Epic Registry" Edition 
MonitorPowerCase
ASUS 21.5 1920x1080 2ms $135 CORSAIR HX850 $120 Mother Earth $free 
  hide details  
Reply
    
CPUMotherboardGraphicsGraphics
Intel Core i5 2500K P8P67 PRO NVIDIA GeForce GTX 470 NVIDIA GeForce GTX 470 
GraphicsRAMRAMRAM
NVIDIA GeForce 9800 GT G-Skill A-Data G-Skill 
RAMHard DriveOptical DriveOS
A-Data Crucial M4 64GB + 1TB F3 Spinpoint $155 LS/DL DVD RW $?? Windows 8 64-bit "Epic Registry" Edition 
MonitorPowerCase
ASUS 21.5 1920x1080 2ms $135 CORSAIR HX850 $120 Mother Earth $free 
  hide details  
Reply
post #20 of 23
Thread Starter 
Awesome news, now I just need money for a new psu and a second 560.

I keep my CPU clocked that high because I convert videos quite frequently. I quick test I did before with my 945 and the 1090t I had for a day. A 90min mkv file two pass converted to divx hd took 50min with the 945. It took 47min with the 1090t, which is why I returned it. The other day, the 2500k took 29min clocked at 4.7ghz
My System
(13 items)
 
  
CPUMotherboardGraphicsRAM
2500k - 4.8ghz @1.35v p67a-g45 MSI gtx560ti 2x2 patriot viper 1600mhz 
Hard DriveOSMonitorPower
3 WD's - 1.7TB Total Win7 Ultimate ViewSonic 2702@1920x1080 & Mitsu 65" @1920x1080 BFG ls550 
Case
800D Obsidian 
  hide details  
Reply
My System
(13 items)
 
  
CPUMotherboardGraphicsRAM
2500k - 4.8ghz @1.35v p67a-g45 MSI gtx560ti 2x2 patriot viper 1600mhz 
Hard DriveOSMonitorPower
3 WD's - 1.7TB Total Win7 Ultimate ViewSonic 2702@1920x1080 & Mitsu 65" @1920x1080 BFG ls550 
Case
800D Obsidian 
  hide details  
Reply
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: NVIDIA
Overclock.net › Forums › Graphics Cards › NVIDIA › MSI GTX 560ti TwinFrozr II: Phenom x4 945 vs. i5-2500k