Overclock.net › Forums › Graphics Cards › AMD/ATI › I'm looking to buy the new 7850 but I'm thinking my Q6600 is gonna just bottleneck it. Help.
New Posts  All Forums:Forum Nav:

I'm looking to buy the new 7850 but I'm thinking my Q6600 is gonna just bottleneck it. Help. - Page 4

post #31 of 38
Spend $250 on a graphics card.

OR

Spend $380 on a motherboard + CPU.

Guess which nets the bigger performance gain? If you can only choose one of the 2... the magical i5-2500k won't give you amazing results with a 4850, I can say that much.

Numbers don't lie. Anand's reference system uses a GTX 580 for the graphics benchmarks - which is a solid 25% faster (not that you can factor that in and extrapolate, except know that percentage-wise, the performance gap will be smaller with cards weaker than the 580). http://www.anandtech.com/bench/Product/50?vs=288

And hey, look at this: 4890 gets 20FPS in BF3, 6950 (slightly slower than 7850) gets almost triple the framerate. So even if the 2500k would push it 25% higher - ie you'd get 45FPS instead of 20 FPS, no big deal! http://www.techspot.com/review/458-battlefield-3-performance/page5.html
 
Lanbox Lite
(16 items)
 
 
CPUGraphicsRAMHard Drive
i3-2310M Intel HD 3000 8GB DDR3 Samsung 830 64GB SSD 
Hard DriveOSMonitorCase
640GB Hitachi HDD Windows 7 13.3" LCD Magnesium Alloy, 3.2lbs 
CPUMotherboardGraphicsRAM
Phenom II X4 955 BE Gigabyte GA-MA78GPM-UD2H MSI Hawk R5770 3x2GB G.Skill DDR2 800 4-4-4-12 
Hard DriveHard DriveOptical DriveCooling
1TB Samsung Spinpoint F3 250 GB WD Caviar Black Samsung 20X DVD-R/RW Thermaltake MaxOrb 
CoolingOSMonitorPower
Noctua NF-B9-1600 Windows 7 Pro 64-bit BenQ E2420HD, 24" 1920x1080 TT Purepower 500W 
Case
TT Lanbox Lite 
  hide details  
Reply
 
Lanbox Lite
(16 items)
 
 
CPUGraphicsRAMHard Drive
i3-2310M Intel HD 3000 8GB DDR3 Samsung 830 64GB SSD 
Hard DriveOSMonitorCase
640GB Hitachi HDD Windows 7 13.3" LCD Magnesium Alloy, 3.2lbs 
CPUMotherboardGraphicsRAM
Phenom II X4 955 BE Gigabyte GA-MA78GPM-UD2H MSI Hawk R5770 3x2GB G.Skill DDR2 800 4-4-4-12 
Hard DriveHard DriveOptical DriveCooling
1TB Samsung Spinpoint F3 250 GB WD Caviar Black Samsung 20X DVD-R/RW Thermaltake MaxOrb 
CoolingOSMonitorPower
Noctua NF-B9-1600 Windows 7 Pro 64-bit BenQ E2420HD, 24" 1920x1080 TT Purepower 500W 
Case
TT Lanbox Lite 
  hide details  
Reply
post #32 of 38
Quote:
Originally Posted by Diamondhead View Post

Quote:
jason387...
"Oh really? If you have a amd quad core processor and you get a certain fps with your gpu usage at 99% how is your card by any means bottle-necked? Please OP have a try,open up msi afterburner and start up battlefield 3 and play it for a while and then have a look at your gpu usage and if the max recorded gpu usage is 99% then you are not gpu bottle-necked as your gpu is running at it's full usage. If however,the usage is low then i can conclude by saying that your cpu does not have enough power to run your gpu at max usage. So we aren't looking at a system which is likely to get more fps in a game,we are looking for the maximum fps contributed by the card. Everybody is aware the games use both-cpu and gpu,some games use more of the cpu than gpu while others the other way around. So both the gpu as well as cpu contribute to the fps you get in a game. Guess in these days the term 'bottleneck' has changed to mean any processor that generates higher fps rather than it traditionally being applied to not enough cpu power to run the gpu at max usage"
Quote:
ronnin426850...
"If a new CPU gives you FPS raise, then by definition, the old CPU was bottlenecking your GPU, and it's usage with the old CPU CAN'T be 99%.
If the GPU usage was 99%, a new CPU will NOT give you FPS raise, use your head for a change"
I have to side with jason387 on this one, and here's why...
I decided to benchmark my sig rig with dual HD 6850's and an AMD Phenom II X4 830 @ 3.8 GHz.
So I ran 3DMark 11, and I received a score of P6698, and according to MSI Afterburner my gpu useage was 99%-98% on both cards throughout the test. Curious, I compared my results to an i3 2100, in the results section.
I found my total score was a couple hundrad points higher then the top ranked HD 6850 Crossfire i3 2100 rig ( which I noticed, had the exact same gpu overclock as well ).
When I did a side by side comparison of the results, I was suprised to find that every GPU test the i3 2100 had me beat between 1-2 fps,
however I had manged to squeak out a victory because I had beaten the I3 2100 by 3-4 fps in both the CPU physics, and combined test.
That got me askin'... If my CPU was able to out muscle the the i3 2100 in both the physics, and combined tests, why didn't that carry over to win in the GPU tests as well? If I had near perfect gpu utilization, how did the i3 system beat me in every single GPU test? Are my GPU's being bottlenecked by my CPU???... My conclusion was, NO my Phenom II X4 830 is not bottleneckin' my GPU's, since I'm getting near perfect GPU utilization. I chalk the difference up to just a difference in CPU architectures, one is better at physics, while the other is better at pushing frame rates... That's all. thumb.gif
Cheers cheers.gif

We're talking real life gaming FPS.
Also, 1-2 FPS could easily be the difference between "near perfect" GPU utilization and perfect GPU utilization. So you still could have had a bottleneck, however minimal.
But again, synthetic benchmarks are not very good in determinimg that kind of stuff, try the Crysis bench instead, or Metro maybe.. smile.gif

Next, your higher CPU and combined score is due to the fact that you compare quad to dual. Those tests are highly core-dependant. The GPU test is not. And core to core, the i3 beats the Phenom II. That is why i3 has better GPU score (because GPU score does not rely so heavily on core count) and lower CPU score (because it depends on core count).
So, since most games are single or dual-core (that is changing fast, so no worries), your GPUs were bottlenecked by the lower core by core performance of the Phenom quad.
If you test a game that is optimized for quads or more, you won't see a bottleneck. But if you test a sinlge or dual-threaded game, the bottleneck is apparent, compared to an i3 smile.gif
My Rig
(14 items)
 
Ex-wife's Rig
(15 items)
 
 
CPUMotherboardGraphicsRAM
Core i5 4460 AsRock H81M-DG4 Sapphire Rx470 Platinum KVR 1600 16Gb 
Hard DriveHard DriveCoolingOS
2x Seagate 3Tb Samsung 850 EVO 120 Scythe Ninja 3 Rev.B Windows 10 Pro 
MonitorKeyboardPowerCase
Fujitsu Siemens A17-2A Logitech K280e SuperFlower SF-550K12XP Thermaltake Versa H25 
MouseAudio
Logitech G402 Sony MDR XD150 
CPUMotherboardGraphicsRAM
Athlon 750K 4.0Ghz AsRock FM2A75 Pro4+ Sapphire R9 270X Dual-X Kingston 2x4Gb 1600 
Hard DriveHard DriveOptical DriveCooling
Samsung 850 EVO 120  Western Digital 320Gb LiteON DVD-RW CoolerMaster Hyper Z600 
OSMonitorKeyboardPower
Windows 7 Pro x64 Toshiba 32" FullHD TV Logitech FSP Hexa 550 
CaseMouse
DeLUX Logitech 
  hide details  
Reply
My Rig
(14 items)
 
Ex-wife's Rig
(15 items)
 
 
CPUMotherboardGraphicsRAM
Core i5 4460 AsRock H81M-DG4 Sapphire Rx470 Platinum KVR 1600 16Gb 
Hard DriveHard DriveCoolingOS
2x Seagate 3Tb Samsung 850 EVO 120 Scythe Ninja 3 Rev.B Windows 10 Pro 
MonitorKeyboardPowerCase
Fujitsu Siemens A17-2A Logitech K280e SuperFlower SF-550K12XP Thermaltake Versa H25 
MouseAudio
Logitech G402 Sony MDR XD150 
CPUMotherboardGraphicsRAM
Athlon 750K 4.0Ghz AsRock FM2A75 Pro4+ Sapphire R9 270X Dual-X Kingston 2x4Gb 1600 
Hard DriveHard DriveOptical DriveCooling
Samsung 850 EVO 120  Western Digital 320Gb LiteON DVD-RW CoolerMaster Hyper Z600 
OSMonitorKeyboardPower
Windows 7 Pro x64 Toshiba 32" FullHD TV Logitech FSP Hexa 550 
CaseMouse
DeLUX Logitech 
  hide details  
Reply
post #33 of 38
get the 7850 it will last you so long you will have upgraded to a new platform anyways. i bought my 4850 back in 09 then got another one recently for 50 bucks crossfired and BOOM big increase and eventually i may get a 7850.
Core 2 Haf
(16 items)
 
  
CPUMotherboardGraphicsRAM
C2Q Q9550 ASUS P5E3 Deluxe Wifi-AP GTX 480 Kingston HyperX DDR3 KHX1600C9D3K2/8G 
RAMHard DriveCoolingOS
PNY 1333 55gb, 320Gb, 500GB, 1TB Hyper 212 EVO Win 7 Ultimate 64bit 
OSOSPowerCase
Lubuntu 13.04 x64 OS X 10.8.5 TX750 Haf 912 
  hide details  
Reply
Core 2 Haf
(16 items)
 
  
CPUMotherboardGraphicsRAM
C2Q Q9550 ASUS P5E3 Deluxe Wifi-AP GTX 480 Kingston HyperX DDR3 KHX1600C9D3K2/8G 
RAMHard DriveCoolingOS
PNY 1333 55gb, 320Gb, 500GB, 1TB Hyper 212 EVO Win 7 Ultimate 64bit 
OSOSPowerCase
Lubuntu 13.04 x64 OS X 10.8.5 TX750 Haf 912 
  hide details  
Reply
post #34 of 38
Allright, the Q6600 will bottleneck a HD7850. That's just plain simple and straight forward.
Allright, you also won't notice it at all. Another simple fact.
You will notice a increase in FPS when you later on switch to a faster CPU.
You will notice a big difference when OCing. Especially if you push well beyond the 3Ghz limit.

Should you buy a HD7850? Yep. No doubt. Is the Q6600 enough to enjoy it in BF3?
Yeah, only it will give you drops in 64 player busy situations and it will give larger drops in 64 player B2K maps as they are slightly heavier to run.

I also get a hard bottleneck with my 1100T @ 4.15Ghz and a CF HD6950 unlocked setup. Strong drops and low GPU usage in 64 player servers.
Got a 2500K laying around able to do 4.9Ghz 24/7 and I will be using it when my new Z77A GD-65 board is in. Once the i7 3770k hits retail i''ll be getting that. That oughtta remove my bottlenecks once and for all.
Thundahcat.
(20 items)
 
   
CPUMotherboardGraphicsRAM
AMD Athlon XP-M Barton @ 2612Mhz ASUS A7N8X-X HIS Radeon X800XL 256MB w/ V-MOD @ 585Mhz GPU 5... 3x512MB Samsung PC-3200 CL3 
Hard DriveCoolingOSOS
Maxtor 80GB IDE Cooler Master alu/copper block with a Slipstrea... Windows 98 SE Windows XP Pro SP3 
PowerCase
Cooler Master Real Power M520W OEM with great airflow and even cable management 
  hide details  
Reply
Thundahcat.
(20 items)
 
   
CPUMotherboardGraphicsRAM
AMD Athlon XP-M Barton @ 2612Mhz ASUS A7N8X-X HIS Radeon X800XL 256MB w/ V-MOD @ 585Mhz GPU 5... 3x512MB Samsung PC-3200 CL3 
Hard DriveCoolingOSOS
Maxtor 80GB IDE Cooler Master alu/copper block with a Slipstrea... Windows 98 SE Windows XP Pro SP3 
PowerCase
Cooler Master Real Power M520W OEM with great airflow and even cable management 
  hide details  
Reply
post #35 of 38
Quote:
Originally Posted by Diamondhead View Post

Quote:
jason387...
"Oh really? If you have a amd quad core processor and you get a certain fps with your gpu usage at 99% how is your card by any means bottle-necked? Please OP have a try,open up msi afterburner and start up battlefield 3 and play it for a while and then have a look at your gpu usage and if the max recorded gpu usage is 99% then you are not gpu bottle-necked as your gpu is running at it's full usage. If however,the usage is low then i can conclude by saying that your cpu does not have enough power to run your gpu at max usage. So we aren't looking at a system which is likely to get more fps in a game,we are looking for the maximum fps contributed by the card. Everybody is aware the games use both-cpu and gpu,some games use more of the cpu than gpu while others the other way around. So both the gpu as well as cpu contribute to the fps you get in a game. Guess in these days the term 'bottleneck' has changed to mean any processor that generates higher fps rather than it traditionally being applied to not enough cpu power to run the gpu at max usage"
Quote:
ronnin426850...
"If a new CPU gives you FPS raise, then by definition, the old CPU was bottlenecking your GPU, and it's usage with the old CPU CAN'T be 99%.
If the GPU usage was 99%, a new CPU will NOT give you FPS raise, use your head for a change"
I have to side with jason387 on this one, and here's why...
I decided to benchmark my sig rig with dual HD 6850's and an AMD Phenom II X4 830 @ 3.8 GHz.
So I ran 3DMark 11, and I received a score of P6698, and according to MSI Afterburner my gpu useage was 99%-98% on both cards throughout the test. Curious, I compared my results to an i3 2100, in the results section.
I found my total score was a couple hundrad points higher then the top ranked HD 6850 Crossfire i3 2100 rig ( which I noticed, had the exact same gpu overclock as well ).
When I did a side by side comparison of the results, I was suprised to find that every GPU test the i3 2100 had me beat between 1-2 fps,
however I had manged to squeak out a victory because I had beaten the I3 2100 by 3-4 fps in both the CPU physics, and combined test.
That got me askin'... If my CPU was able to out muscle the the i3 2100 in both the physics, and combined tests, why didn't that carry over to win in the GPU tests as well? If I had near perfect gpu utilization, how did the i3 system beat me in every single GPU test? Are my GPU's being bottlenecked by my CPU???... My conclusion was, NO my Phenom II X4 830 is not bottleneckin' my GPU's, since I'm getting near perfect GPU utilization. I chalk the difference up to just a difference in CPU architectures, one is better at physics, while the other is better at pushing frame rates... That's all. thumb.gif
Cheers cheers.gif

At last a test was done. This is actually the best way to go about solving it:)
Jay
(13 items)
 
  
CPUMotherboardGraphicsRAM
FX 8350 4.4Ghz at 1.36v GA 78LMT USB3 Rev 5.0 Sapphire R9 285 OC 16GB Kingston Savage Ram 
Hard DriveHard DriveCoolingOS
1TB Western Digital 500GB Seagate Barracuda Noctua NH-D15S Windows 8.1 
MonitorPowerMouseMouse Pad
Asus MG279Q Fressync OCZ ZS 650 Watt PSU Dragonwar ELE-G9 Dragonwar 
  hide details  
Reply
Jay
(13 items)
 
  
CPUMotherboardGraphicsRAM
FX 8350 4.4Ghz at 1.36v GA 78LMT USB3 Rev 5.0 Sapphire R9 285 OC 16GB Kingston Savage Ram 
Hard DriveHard DriveCoolingOS
1TB Western Digital 500GB Seagate Barracuda Noctua NH-D15S Windows 8.1 
MonitorPowerMouseMouse Pad
Asus MG279Q Fressync OCZ ZS 650 Watt PSU Dragonwar ELE-G9 Dragonwar 
  hide details  
Reply
post #36 of 38
Thread Starter 
Thanks everyone! It looks like it really is work it! I'm going to try to get it soon and eventually upgrade the rest, but for now all I can do is this. Thanks again! smile.gif
post #37 of 38
I think the main bottleneck issue is not being discussed properly.
The FPS performance depends on the GPU and CPU in different proportions depending on the code of the app and many other factors.
A perfect code would always be able to use 100% of both CPU and GPU (without frame limiting) , however, nowadays this is hardly achievable specially with ported games.
There are many stories about poorly optimized PC ports (GTAIV) and whole engines (Fallout 3, Skyrim).
I think the CPU can limit performance at certain degrees if it can't cope with GPU calculations in a routine, that would be CPU 100% GPU <100 % which is just theoretical.
For that I use as an example the FurMark benchmark, that varies little depending on the CPU, specially at higher resolutions. Other apps, specially the ones that use physics and more CPU routines can be even bottlenecked by a 2600k.

My answer is: The degree of bottleneck depends heavily on the application. I think the Core2Quad architecture is still very viable at least for the HD7xxx series since these CPU perform close to the top notch AMD ones.
Ztylo
(17 items)
 
  
CPUMotherboardGraphicsRAM
Q8200 MSI P35 Neo2 FR Sapphire HD 7870XT 4 GB (2x1GB Mushkin + 1x2GB OCZ) 
Hard DriveHard DriveHard DriveOptical Drive
Hitachi HDS7200 Hitachi HDS7200 OCZ Vertex Samsung Writemaster DVDRW 
CoolingOSMonitorKeyboard
Thermaltake TMGi1 Windows 8 x64 Viewsonic VX2250 CMStorm Quickfire Rapid CherryMX Brown 
PowerCaseMouseMouse Pad
Antec Earthwatts Platinum 650W Sunbeamtech VentMax CMStorm Spawn // SSeries WoW Gold Puretrack Stealth // SSeries QCK 
Audio
Xonar D1 
  hide details  
Reply
Ztylo
(17 items)
 
  
CPUMotherboardGraphicsRAM
Q8200 MSI P35 Neo2 FR Sapphire HD 7870XT 4 GB (2x1GB Mushkin + 1x2GB OCZ) 
Hard DriveHard DriveHard DriveOptical Drive
Hitachi HDS7200 Hitachi HDS7200 OCZ Vertex Samsung Writemaster DVDRW 
CoolingOSMonitorKeyboard
Thermaltake TMGi1 Windows 8 x64 Viewsonic VX2250 CMStorm Quickfire Rapid CherryMX Brown 
PowerCaseMouseMouse Pad
Antec Earthwatts Platinum 650W Sunbeamtech VentMax CMStorm Spawn // SSeries WoW Gold Puretrack Stealth // SSeries QCK 
Audio
Xonar D1 
  hide details  
Reply
post #38 of 38
Quote:
Originally Posted by nachomaster View Post

Thanks everyone! It looks like it really is work it! I'm going to try to get it soon and eventually upgrade the rest, but for now all I can do is this. Thanks again! smile.gif

Did you buy 7850? I want to buy it too. I have similar cofiguration smile.gif

A very important thing for oc q6600

I overclock my q6600 at 3.8 ghz , 1.63v . I first started prime95 and let him to work 30 minutes, and everything was ok. After pirme I started Inetl Burn Test, and my pc rapidly turns off after 5 sec, and after that I got black screen, so I had to pull out the cable from the power supply for a while and return it back(standard procedure for asus motherboard when you screw something in bios). If I raise the voltage, I think it just shuts down faster, and If I reduce it I get a normal restet or freezing. I find that it is stable in Intel Burn test on 3.6ghz at 1.55v. But if I leave buss speed at same settings and if I raise voltage I got the same rapid turns off. I followed the temperature of cpu with aida64 extreme and everything was ok, cpu in full load never goes over 61c, ( I am using noctua nh-d14).Many people thought it was a overheating CPU, but I didn't believe in that, because I think that is not possible that cpu hit 100c in few sec with noctua smile.gif So I think that he lacks power...

I bought new pcu , and add more power to cpu , 4pin+4pin ( I didn't have another 4pin of 12v on my old pcu ) , and now works great. I stabilized q6600 at 3.8 with 1.55v(Vdrop is about 0.1v, I set in bios 1.65 ) , and it passed Intel burn test with max temp 78c with Noctua D-14 It is very good, because I didn't think that is possible on 3.8ghz smile.gif . In idle temp is about 31-32 at 25c ambient temp.

Prime 95 is a Piece of Cake for Intel Burn Test smile.gif IntelBurnTest easily heats up Q6600's more than 22*C higher than Prime95,
Edited by delevic - 4/16/12 at 10:33am
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: AMD/ATI
Overclock.net › Forums › Graphics Cards › AMD/ATI › I'm looking to buy the new 7850 but I'm thinking my Q6600 is gonna just bottleneck it. Help.