Overclock.net › Forums › Graphics Cards › AMD/ATI › cpu -GPU Bottleneck?!
New Posts  All Forums:Forum Nav:

cpu -GPU Bottleneck?! - Page 3

post #21 of 38
Yes, AMD can be a good cpu for gaming but is it better than intel- NO! I don't have to point out the scores of benchmarks that prove this since you have internet (since you are reading this) and for anyone to claim AMD is a better performer for gaming or video editing is flat wrong. The price war is a completely different story that people can argue all day long about but the fact remains that intel makes a better product at the end of the day.

http://www.overclock.net/t/1160043/top-30-cinebench-r11-5-cpu-scores
White Knight
(0 items)
  
Reply
White Knight
(0 items)
  
Reply
post #22 of 38
Quote:
Thubans are pretty bad overall for gaming

Please explain your summation that has been quoted.
    
CPUMotherboardGraphicsRAM
INTEL ASUS XFX  SAMSUNG 
Hard DriveOptical DriveCoolingOS
WD/ST LG KUHLER WINDOWS 
MonitorKeyboardPowerCase
LG/SAMSUNG IBM MODEL M CORSAIR THERMALTAKE 
MouseMouse PadAudio
MS INTELLIMOUSE EXPLORER 3.0 REGULAR LARGE PAD ONBOARD but it USED TO BE A XONAR DG  
  hide details  
Reply
    
CPUMotherboardGraphicsRAM
INTEL ASUS XFX  SAMSUNG 
Hard DriveOptical DriveCoolingOS
WD/ST LG KUHLER WINDOWS 
MonitorKeyboardPowerCase
LG/SAMSUNG IBM MODEL M CORSAIR THERMALTAKE 
MouseMouse PadAudio
MS INTELLIMOUSE EXPLORER 3.0 REGULAR LARGE PAD ONBOARD but it USED TO BE A XONAR DG  
  hide details  
Reply
post #23 of 38
OP: i vote for just overclocking your 1055t.
dont buy another cpu.
your thuban is GOOD. imo its one of the best amds u can get right now. (at least in terms of value imo)
also for skyrim, try the performance and 4gb mods that helped my fps on skyrim alot.
before i would get anywhere from 50-120 and it fluctuated very frequently.
now its anywhere from 80-120 and fluctuates less frequently.
post #24 of 38
Quote:
Originally Posted by GanjaSMK View Post

Please explain your summation that has been quoted.

I'm assuming you don't have the ability to objectively look at a chart and read numbers I guess...
2500k
(15 items)
 
  
CPUMotherboardGraphicsRAM
2500k @ 4.6 1.27V Asus p67 Pro SLI 560 TI Ripjaws X 1600 2x4GB 
Hard DriveCoolingOSMonitor
Crucial C300 + Spinpoint F3 Noctua N-D14 Win 7 x64 LG 1920x1080 
KeyboardPowerCaseMouse
Das Professional NZXT 750w Hale90 Antec 600 M60 
Audio
ATH AD700 
  hide details  
Reply
2500k
(15 items)
 
  
CPUMotherboardGraphicsRAM
2500k @ 4.6 1.27V Asus p67 Pro SLI 560 TI Ripjaws X 1600 2x4GB 
Hard DriveCoolingOSMonitor
Crucial C300 + Spinpoint F3 Noctua N-D14 Win 7 x64 LG 1920x1080 
KeyboardPowerCaseMouse
Das Professional NZXT 750w Hale90 Antec 600 M60 
Audio
ATH AD700 
  hide details  
Reply
post #25 of 38
Quote:
Originally Posted by quesoblanco View Post

yes u r bottlenecking. On bf3 my gpu usage was utter trash. I was rockin a 1090t 4.0 OC 2600cpu/nb and it could not keep up with a 6850 cf set up. my cpu even with vsync on was constant 90 percent and my gpu usage was like 60-70 percent. Get an i5 2500k and itll take care of your problems. If u must stick with ur thuban then sell ur 6850s and buy one good card. Itll relieve the bottleneck by a lot or hopefully get rid of it.... It also depends on what games you play. In battlefield games, ull be bottlenecked.

i hope your GPU usage is better than your posts.
post #26 of 38
Grief - the PM to you:
Quote:
Listen here. I have a problem with many people portraying the Phenom II lineup, including the Thubans, as being (my own quote) 'inadequate' for gaming. This is just a silly, false, and misguided fallacy.

There is reason to show that some games don't use more than two cores - everyone gets that.
There is reason to show that some games are CPU bound - everyone gets that.
There is also reason to show that some games are GPU bound - and everyone gets that.

What I don't get is running up charts and benchmarks that don't effectively portray the whole picture surrounding a game. Did the reviewer overclock anything? What about the CPU/NB if it was an Phenom II? What about real-world usage instead of benchmarks? Did they overclock via a BE multiplier or did they use HTT, or a combination of both? Are the systems completely identical beyond differences in hardware?

My point is - I've played SC2, BFBC2, COD4, Oblivion, Just Cause 2, Batman AA, Crysis, Crysis Warhead, Crysis 2, BF3, MoH (2010), Brink, Defense Grid, AC, ACII, Metro 2033, Battleforge, LoL, Dota, SC:C and hosts more of other 'relevant' games that everyone plays and never once has my Phenom II X3, X4, or X6 disappointed in performance. Of course all of the chips were pushed to their highest clocks and yet I still don't find my numbers matching what other people find in comparison to benchmarks. But even beyond my personal experiences there are several factors that come into play:

* Player expectations
* Game performance expectations
* Actual performance
* Frame rate often dependent upon game
* Fluidity of motion to the individual playing
* Stability of an overclocked system
* Driverbase
* Margins of error


On top of that you claim that "Thubans are pretty bad overall for gaming" when it simply isn't true. Now let's head to the beast in the room. Sure, Intel has been shown t outperform AMD in several games with different chips. In fact some of the Core2 chips scored better than some of the Thubans in some testing scenarios! Wow, so that Thuban must be a horrible horrible chip to contemplate getting because it failed some synthetic benchmarks, right? Not so - everyone knows first and second gen i-Series processors have been out and around for some time, as have the Thubans, and everyone knows they are definitely superior. It doesn't make the Thubans trash for gaming.

But in reality - the person that bought the Thuban might not have wanted to switch full setups - perhaps they had an Athlon X2 or X3. Or a Phenom II 710. Perhaps it was a drop-in replacement instead of a complete platform upgrade. Sometimes that just makes sense. And they do perform. They're perfectly fine for gaming, multitasking, daily use, and otherwise.

I mean here, if you want it - here it is:
Quote:
700
677

And here we go - lets really look at this:
Quote:
And I'll note - the minimums are righteously close between the FX, 1100T, and the 2500K with a dip at 10 FPS towards the 980, in the second part of your quote, where the settings are on 'Medium'. The minimums on the top part - when the settings are at 'Ultra' show exactly the same kind of relation, FX, 1100T, 2500K all roughly the same, with a 10 FPS dip towards the 980. 10 whole frames...

That make make a difference for sure if you're jumping from 20 frames for a minimum to 30, but come on now. That's a horrible excerpt to be quoting on behalf of what you state. You having upgraded to a 2500K to solve your problem was your choice and no one including myself, is arguing against that.

Just looking some more - it's even more relevant to note - with a 6990, the world's 'fastest' video card, there is a mere 12 frames of difference in minimum FPS between the 2600K and the 980. 12 frames. If that doesn't tell you the game is much more GPU bound than CPU bound, noting that the X6 from Intel in that test shows only a single frame higher minimum than the Intel top quad, I don't know what will. (Me from this thread

So in conclusion, please advise me how an Thuban is 'pretty bad overall for gaming', when evidence would completely suggest otherwise?
    
CPUMotherboardGraphicsRAM
INTEL ASUS XFX  SAMSUNG 
Hard DriveOptical DriveCoolingOS
WD/ST LG KUHLER WINDOWS 
MonitorKeyboardPowerCase
LG/SAMSUNG IBM MODEL M CORSAIR THERMALTAKE 
MouseMouse PadAudio
MS INTELLIMOUSE EXPLORER 3.0 REGULAR LARGE PAD ONBOARD but it USED TO BE A XONAR DG  
  hide details  
Reply
    
CPUMotherboardGraphicsRAM
INTEL ASUS XFX  SAMSUNG 
Hard DriveOptical DriveCoolingOS
WD/ST LG KUHLER WINDOWS 
MonitorKeyboardPowerCase
LG/SAMSUNG IBM MODEL M CORSAIR THERMALTAKE 
MouseMouse PadAudio
MS INTELLIMOUSE EXPLORER 3.0 REGULAR LARGE PAD ONBOARD but it USED TO BE A XONAR DG  
  hide details  
Reply
post #27 of 38
Quote:
Originally Posted by GanjaSMK View Post

Grief - the PM to you:

Wow...that was a long message with really nothing in it.

Basically all you are saying is you have played a lot of games, core 2's beat thubans, and some people need to stick with with the same socket.

I understand these points, but I don't see how in anyway this supports your argument...you really aren't saying anything if you break it down.

Beyond that, as far as I can see your only two arguments are benchmarks aren't relevant because they are at stock clocks and 'hey look, bf3 benchmarks.

I think it is hilarious that you linked benchmarks, since you spent half of your post arguing about how pointless and irrelevant they are.
Quote:
Originally Posted by GanjaSMK View Post

Wow, so that Thuban must be a horrible horrible chip to contemplate getting because it failed some synthetic benchmarks, right?


What I don't get is running up charts and benchmarks that don't effectively portray the whole picture surrounding a game. Did the reviewer overclock anything? What about the CPU/NB if it was an Phenom II? What about real-world usage instead of benchmarks? Did they overclock via a BE multiplier or did they use HTT, or a combination of both? Are the systems completely identical beyond differences in hardware?

But I have also mentioned Battlefield 3 benchmarks on the first page...so I don't see why you think that was necessary...

Quote:
Originally Posted by Grief View Post

Most games aren't multi-threaded, bf3 has a pc optimized engine though and scales well with multiple cores...99% of games won't.

Edited by Grief - 12/11/11 at 12:02pm
2500k
(15 items)
 
  
CPUMotherboardGraphicsRAM
2500k @ 4.6 1.27V Asus p67 Pro SLI 560 TI Ripjaws X 1600 2x4GB 
Hard DriveCoolingOSMonitor
Crucial C300 + Spinpoint F3 Noctua N-D14 Win 7 x64 LG 1920x1080 
KeyboardPowerCaseMouse
Das Professional NZXT 750w Hale90 Antec 600 M60 
Audio
ATH AD700 
  hide details  
Reply
2500k
(15 items)
 
  
CPUMotherboardGraphicsRAM
2500k @ 4.6 1.27V Asus p67 Pro SLI 560 TI Ripjaws X 1600 2x4GB 
Hard DriveCoolingOSMonitor
Crucial C300 + Spinpoint F3 Noctua N-D14 Win 7 x64 LG 1920x1080 
KeyboardPowerCaseMouse
Das Professional NZXT 750w Hale90 Antec 600 M60 
Audio
ATH AD700 
  hide details  
Reply
post #28 of 38
Quote:
Originally Posted by Grief View Post

Wow...that was a long message with really nothing in it.
Basically all you are saying is you have played a lot of games, core 2's beat thubans, and some people need to stick with with the same socket.
I understand these points, but I don't see how in anyway this supports your argument...you really aren't saying anything if you break it down.
Beyond that, as far as I can see your only two arguments are benchmarks aren't relevant because they are at stock clocks and 'hey look, bf3 benchmarks.
I think it is hilarious that you linked benchmarks, since you spent half of your post arguing about how pointless and irrelevant they are.
But I have also mentioned Battlefield 3 benchamarks on the first page...so I don't see why you think that was necessary...

You may take it how you perceive it, that's a right given to any human. And if you perceive it the way you do, I can equally say in contrast that your benchmarks show no relevance to support your argument. I'm not making an argument by the way, I'm simply trying to understand how, "Thubans are pretty bad overall for gaming".

And just to point out, I never said nor implied that benchmarks were pointless or irrelevant. Again your perception is playing a role here.

If you can't explain why "Thubans are pretty bad overall for gaming", or point out where this is true I certainly understand. No harm no foul. thumb.gif
    
CPUMotherboardGraphicsRAM
INTEL ASUS XFX  SAMSUNG 
Hard DriveOptical DriveCoolingOS
WD/ST LG KUHLER WINDOWS 
MonitorKeyboardPowerCase
LG/SAMSUNG IBM MODEL M CORSAIR THERMALTAKE 
MouseMouse PadAudio
MS INTELLIMOUSE EXPLORER 3.0 REGULAR LARGE PAD ONBOARD but it USED TO BE A XONAR DG  
  hide details  
Reply
    
CPUMotherboardGraphicsRAM
INTEL ASUS XFX  SAMSUNG 
Hard DriveOptical DriveCoolingOS
WD/ST LG KUHLER WINDOWS 
MonitorKeyboardPowerCase
LG/SAMSUNG IBM MODEL M CORSAIR THERMALTAKE 
MouseMouse PadAudio
MS INTELLIMOUSE EXPLORER 3.0 REGULAR LARGE PAD ONBOARD but it USED TO BE A XONAR DG  
  hide details  
Reply
post #29 of 38
Intel's current cpu's are better than current AMD cpu's, in gaming, but generally in anything. But that doesn't turn the amd cpu you currently have into crap. You haven't really said if performance in BF3 is disappointing, but at stock your cpu is probably bottlenecking yes. It certainly couldn't hurt to overclock it to 3.7GHz at least. Although I don't know what cooler you have on it right now I'm certain you know what things to look at while overclocking. (else you could always ask of course)
Isn't BF3 actually one of the few games in which the thuban cpu's are noticeably faster than their deneb counterparts? Anyway, crossfire support for Skyrim isn't... that great. Let's just say that measuring Skyrim performance isn't a good way to spot bottlenecks or measure system performance atm.
Ya-1800
(14 items)
 
  
CPUMotherboardGraphicsRAM
Ryzen 7 1800X ASRock X370 Taichi Sapphire Radeon R9 290 G.Skill Flare X F4-3200C14D-16GFX 
Hard DriveHard DriveHard DriveHard Drive
Samsung 840 Pro 256GB Samsung 850 EVO 1TB Samsung HD103SJ 1TB Samsung HD204UI 2TB 
Hard DriveCoolingKeyboardPower
Western Digital RED WD40EFRX 4 TB EK Supremacy EVO Filco Majestouch 2 PC Power&Cooling 750 Watt 
CaseMouse
Phanteks Enthoo Primo M570 Trackball 
  hide details  
Reply
Ya-1800
(14 items)
 
  
CPUMotherboardGraphicsRAM
Ryzen 7 1800X ASRock X370 Taichi Sapphire Radeon R9 290 G.Skill Flare X F4-3200C14D-16GFX 
Hard DriveHard DriveHard DriveHard Drive
Samsung 840 Pro 256GB Samsung 850 EVO 1TB Samsung HD103SJ 1TB Samsung HD204UI 2TB 
Hard DriveCoolingKeyboardPower
Western Digital RED WD40EFRX 4 TB EK Supremacy EVO Filco Majestouch 2 PC Power&Cooling 750 Watt 
CaseMouse
Phanteks Enthoo Primo M570 Trackball 
  hide details  
Reply
post #30 of 38
Quote:
Originally Posted by GanjaSMK View Post

You may take it how you perceive it, that's a right given to any human. And if you perceive it the way you do, I can equally say in contrast that your benchmarks show no relevance to support your argument. I'm not making an argument by the way, I'm simply trying to understand how, "Thubans are pretty bad overall for gaming".
And just to point out, I never said nor implied that benchmarks were pointless or irrelevant. Again your perception is playing a role here.
If you can't explain why "Thubans are pretty bad overall for gaming", or point out where this is true I certainly understand. No harm no foul. thumb.gif

Thubans are pretty bad overall for gaming

Most games aren't multi-threaded (see: just about every game except BF3)
Thubans are based off of dated architecture, and the clock per core will be lower then other processors of similar architecture (see: 955 line)
At stock clocks, and unless you OC it fairly high will cause a bottleneck on newer generation cards in SLI or crossfire. Whether through inability to overclock, lack of knowledge on the subject, or just a bad piece of silicon, you might not be able to use your computer to the fullest potential.

I mean I've pretty much said all of these things multiple times, and the third was implied by the thread and many like it I've seen on OCN and other sites.
2500k
(15 items)
 
  
CPUMotherboardGraphicsRAM
2500k @ 4.6 1.27V Asus p67 Pro SLI 560 TI Ripjaws X 1600 2x4GB 
Hard DriveCoolingOSMonitor
Crucial C300 + Spinpoint F3 Noctua N-D14 Win 7 x64 LG 1920x1080 
KeyboardPowerCaseMouse
Das Professional NZXT 750w Hale90 Antec 600 M60 
Audio
ATH AD700 
  hide details  
Reply
2500k
(15 items)
 
  
CPUMotherboardGraphicsRAM
2500k @ 4.6 1.27V Asus p67 Pro SLI 560 TI Ripjaws X 1600 2x4GB 
Hard DriveCoolingOSMonitor
Crucial C300 + Spinpoint F3 Noctua N-D14 Win 7 x64 LG 1920x1080 
KeyboardPowerCaseMouse
Das Professional NZXT 750w Hale90 Antec 600 M60 
Audio
ATH AD700 
  hide details  
Reply
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: AMD/ATI
Overclock.net › Forums › Graphics Cards › AMD/ATI › cpu -GPU Bottleneck?!