• Welcome to TechPowerUp Forums, Guest! Please check out our forum guidelines for info related to our community.

The Search for Micro Stuttering in 4870X2

FelipeV

New Member
Joined
Apr 1, 2008
Messages
95 (0.02/day)
Location
Brazil
Processor Core2Duo E8500@4250Mhz 8,5x500
Motherboard ASUS P5Q-E
Cooling Thermal True X 120
Memory 8GB Gskill F2-8000CL5D-4GBPQ @ 1000Mhz
Video Card(s) PowerColor HD 4870X2
Storage 2x320GB R0 + 2x80GB
Display(s) AOC 212va 22"
Case Asys Freedom
Audio Device(s) Onboard
Power Supply Corsair TX750
Software Vista Business 64/ W7 RC7100 64
Me and my friends start testing the new X2 from ATI and comparing it with the 3870X2 and a Single 4870.

This are the first comparisson we make comparing the frametimes between this cards, and we get surprise to see the results of the new X2.

EDIT: We use Fraps to record Frametimes during a benchmark run, those datas where calculated using Excell to compare the time between frametimes in each card. Is that okay or need more detailed info ? I´m not the math guy ;) for that I have to ask for the guy who built the formula hehehe.

Frametimes Explanation.
http://www.mvps.org/directx/articles/fps_versus_frame_time.htm


Y= Miliseconds
X = Frame number


Example of the data collected.


This is DMC4 1680x1050 4xAA.


COJ 1920x1080 4xSSAA


The upcoming tests will be Crysis and Grid.

And, I´m here to ask users with Nvidia SLI configurations if you can help running tests to compare the evolution in MGPU configs. If you are willing to do so, please PM me with so I can send the settings used in the test.

Users with Single cards from NV are welcome too.

And if any of you got any suggestions feel free to say ;)
 
Last edited:

EastCoasthandle

New Member
Joined
Apr 21, 2005
Messages
6,885 (0.99/day)
System Name MY PC
Processor E8400 @ 3.80Ghz > Q9650 3.60Ghz
Motherboard Maximus Formula
Cooling D5, 7/16" ID Tubing, Maze4 with Fuzion CPU WB
Memory XMS 8500C5D @ 1066MHz
Video Card(s) HD 2900 XT 858/900 to 4870 to 5870 (Keep Vreg area clean)
Storage 2
Display(s) 24"
Case P180
Audio Device(s) X-fi Plantinum
Power Supply Silencer 750
Software XP Pro SP3 to Windows 7
Benchmark Scores This varies from one driver to another.
Before posting a bunch of graphs please example how you obtained the data to plot each dot/line
 

FelipeV

New Member
Joined
Apr 1, 2008
Messages
95 (0.02/day)
Location
Brazil
Processor Core2Duo E8500@4250Mhz 8,5x500
Motherboard ASUS P5Q-E
Cooling Thermal True X 120
Memory 8GB Gskill F2-8000CL5D-4GBPQ @ 1000Mhz
Video Card(s) PowerColor HD 4870X2
Storage 2x320GB R0 + 2x80GB
Display(s) AOC 212va 22"
Case Asys Freedom
Audio Device(s) Onboard
Power Supply Corsair TX750
Software Vista Business 64/ W7 RC7100 64

EastCoasthandle

New Member
Joined
Apr 21, 2005
Messages
6,885 (0.99/day)
System Name MY PC
Processor E8400 @ 3.80Ghz > Q9650 3.60Ghz
Motherboard Maximus Formula
Cooling D5, 7/16" ID Tubing, Maze4 with Fuzion CPU WB
Memory XMS 8500C5D @ 1066MHz
Video Card(s) HD 2900 XT 858/900 to 4870 to 5870 (Keep Vreg area clean)
Storage 2
Display(s) 24"
Case P180
Audio Device(s) X-fi Plantinum
Power Supply Silencer 750
Software XP Pro SP3 to Windows 7
Benchmark Scores This varies from one driver to another.
Look there, is that okay ?

those datas where calculated using Excell to compare the time between frametimes in each card
This needs to be explained and shown. That is the heart of the graphs being shown.

-How are you determining ms?
-Where are the ms numbers coming from?
-What settings are used on the card to create the results?
-etc
 
Last edited:
Joined
Nov 14, 2007
Messages
1,773 (0.30/day)
Location
Detroit, MI
Processor AMD Ryzen 7 5800X3D
Motherboard ASUS TUF Gaming X570-Pro Wifi II
Cooling Hyper 212 EVO v2
Memory 2x16GB G.Skill DDR4-4000
Video Card(s) AMD RX 6750 XT
Storage WD Black SN850 2TB, various other SSD's from ages past
Display(s) LG 27GL850 1440@144, AG Neovo EM2701QC 1440@75
Case Zophos EVO Silent by Raijintek
Audio Device(s) HyperX Cloud II Wireless headphones
Power Supply Corsair RM850x
Mouse ProtoArc EM01
Keyboard Razer Blackwidow X Chroma Mercury
Software Windows 11 Pro
I'm a little confused as to what the numbers to the left are... I'd assume the ones on the bottom are seconds?
 
Joined
Feb 14, 2008
Messages
988 (0.17/day)
Location
Cincinnati, Ohio
Processor AMD Ryzen 1700 @ 3.825ghz 1.36v
Motherboard ASRock x370 Taichi
Cooling Noctua NH-C14s
Memory Corsair Vengeance RGB 16gb 3200
Video Card(s) Gigabyte GTX 1080 ti
Display(s) LG 4k IPS
Case Be Quiet Pure Base 600 + more fans
Audio Device(s) Sound Blaster Z
Power Supply EVGA P2 750w
Mouse G900
Keyboard G810
Software Windows 10 Pro
Let me sum it up for everyone: The 4870X2 line is a heck of a lot less bouncy than the 3870X2 line, which means less microstuttering. The 4870 single is shown to show what a single cards "bounciness" is supposed to be at to not notice microstutter.

Did I explain it right? :)
 

FelipeV

New Member
Joined
Apr 1, 2008
Messages
95 (0.02/day)
Location
Brazil
Processor Core2Duo E8500@4250Mhz 8,5x500
Motherboard ASUS P5Q-E
Cooling Thermal True X 120
Memory 8GB Gskill F2-8000CL5D-4GBPQ @ 1000Mhz
Video Card(s) PowerColor HD 4870X2
Storage 2x320GB R0 + 2x80GB
Display(s) AOC 212va 22"
Case Asys Freedom
Audio Device(s) Onboard
Power Supply Corsair TX750
Software Vista Business 64/ W7 RC7100 64
Let me sum it up for everyone: The 4870X2 line is a heck of a lot less bouncy than the 3870X2 line, which means less microstuttering. The 4870 single is shown to show what a single cards "bounciness" is supposed to be at to not notice microstutter.

Did I explain it right? :)


Yes, thats right :rockout:
 

dragonavenger

New Member
Joined
Apr 1, 2008
Messages
6 (0.00/day)
Let me sum it up for everyone: The 4870X2 line is a heck of a lot less bouncy than the 3870X2 line, which means less microstuttering. The 4870 single is shown to show what a single cards "bounciness" is supposed to be at to not notice microstutter.

Did I explain it right? :)


yeahhh:toast:


NewGen Pci-e bridge Rulez!!!
 
Top