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

Memory bandwidth tests... any real differences (part 2)

Joined
Jun 11, 2007
Messages
192 (0.03/day)
Processor i7-3770K @ 45x100
Motherboard P8Z77-V Pro
Cooling Noctua NH-D14
Memory G.SKILL Ripjaws X Series (2 x 8GB) DDR3 1600/F3-1600C9D-16GXM
Video Card(s) Onboard HD4000
Storage Vertex 4 128 GB + other HDDs
Case P183
Power Supply Seasonic SS-560KM
About 7 months ago I posted data comparing two memory dividers (1:1 and 3:5 @ 333 MHz) on my then Q6600/P965 based system and concluded that for the 67 % increase in memory bandwidth, the marginal gains in actual performance weren't worth the extra voltage/heat.

Since then I've upgraded my hardware to an X3360/P35 setup and wanted to revisit this issue. Again, two dividers were looked at: one pair running 8.5x333=2.83 GHz, and another running @ 8.5x400=3.40 GHz:

333 MHz FSB:
1:1 a.k.a. PC2-5300 (667 MHz)
5:8 a.k.a. PC2-8500 (1,067 MHz)

400 MHz FSB:
1:1 a.k.a. PC2-6400 (800 MHz)
4:5 a.k.a. PC2-8000 (1,000 MHz)

I figured there would be a much greater difference in the 333 FSB case since the memory bandwidth increased by 60 % vs. 25 % in the 400 MHz FSB case. All other BIOS settings were held constant with the exception of the divider (and the strap) and the given FSB. Subtimings were set to auto and as such could vary as managed by the board which I found out, was required since manually settings some of the subtimings lead to either an incomplete POST, or an unstable system.

The benchmarks were broken down into three categories:
1) "Real-World" Applications
2) 3D Games
3) Synthetic Benchmarks

The following "real-world" apps were chosen: x264, winrar, and the trial version of Photohop CS3. All were run on a freshly installed version of Windows XP Pro x64 SP2 w/ all relevant hotfixes. The 3D games were just Doom3 (an older game) and Crysis (a newer game). Finally, I threw in some synthetic benchmarks consisting of the Winrar self test, Super Pi-mod, and Everest's synthetic memory benchmark. Here is an explanation of the specifics:

Trial of Photoshop CS3 – The batch function in PSCS3 v10.0.1 was used process a total of fifty-six, 10.1 MP jpeg files (226 MB totally):

1) bicubic resize 10.1 MP to 2.2 MP (3872x2592 --> 1800x1200) which is the perfect size for a 4x6 print @ 300 dpi.
2) smart sharpen (120 %, 0.9 px radius, more accurate, lens blur setting)
3) auto levels
4) saved the resulting files as a quality 10 jpg.

Benchmark results are an average of two runs timed with a stopwatch.

RAR version 3.71 – rar.exe ran my standard backup batch file which generated about 955 MB of rars containing 5,210 files totally. Here is the commandline used:
Code:
rar a -m3 -md4096 -v100m -rv40p -msjpg;mp3;tif;avi;zip;rar;gpg;jpg "f:\Backups\Backup.rar" @list.txt
where list.txt a list of all the target files/dirs included in back up set. Benchmark results are an average of two runs timed with a stopwatch.

x264 Benchmark HD – Automatically runs a 2-pass encode on the same 720p MPEG-2 (1280x720 DVD source) file four times totally. It contains two versions of x264.exe and runs it on both. The benchmark is the best three of four runs (FPS) converted to total encode time.

Shameless promotion --> you can read more about the x264 Benchmark HD at this URL which contains results for hundreds of systems. You can also download the benchmark and test your own machine.

3D Games Based Benchmarks

Doom3 - Ran timeddemo demo1 a total of three times and averaged the fps as the result. Settings were 1,280x1,024, ultra quality with 8x AA.

Crysis - Ran the included "Benchmark_CPU.bat" and "Benchmark_GPU.bat" both of which runs the pre-defined timedemo, looped four times. I took the best three of four (average FPS) and averaged them together as the benchmark. Settings were 1,024x768, very high for all (used the DX9 very high settings hack, and 2x AA.

"Synthetic" Application Based Tests

WinRAR version 3.71 – If you hit alt-B in WinRAR, it'll run a synthetic benchmark. This was run twice (stopped after 150 MB) and is the average of four runs.

SuperPI / mod1.5 XS – The 16M test was run twice, and the average of the two are the benchmark.

Everest v4.50.1330 Memory Benchmark - Ran this benchmark a total of three times and averaged the results.

Hardware specs:
Code:
D.F.I. LP LT P35-TR2 (BIOS: LP35D317)
Intel X3360 @ 8.5x400=3.40 GHz
Corsair Dominator DDR2-1066 (TWIN2X4096-8500C5DF)
   2x 2Gb @ 5-5-5-15 (all subtimings on auto)

 (tRD=8) @ 667 MHz (1:1) @ 2.100V
 (tRD=7) @ 1,066 MHz (5:8) @ 2.100V
 (tRD=8) @ 800 MHz (1:1) @ 2.100V
 (tRD=6) @ 1,000 MHz (4:5) @ 2.100V

EVGA Geforce 8800GTS (G92) w/ 512 meg
Core=770 MHz
Shader=1,923 MHz
Memory=2,000 MHz
Note: the performance levels (tRD) are set automatically by the board which wouldn't POST if I manually tweaked them. Even though they're different, I still feel the data are valid since this is the only way I can run them. In other words, if I'm going to run the higher dividers, it'll be as such or it won't POST!

Without further ado, here are the data starting first with a 333 MHz FSB comparing the 1:1 vs. 5:8 divider (DDR2-667 vs. DDR-1066):


Here are the averaged data visualized graphically:


Now on to the 400 MHz FSB comparing the 1:1 vs. 4:5 divider (DDR2-800 vs. DDR2-1000):


And graphically:


As you can see, there way nothing spectacular in either the real-world category, or the 3D games category in comparison to the massive increase in memory bandwidth (shown on the graphs in red). In fact, I was surprised to see that there were really no gains by Doom3 and minimal gains by Crysis. This is probably due to the fact that the video card shoulders the burden of these games with Doom3 being the light-weight of the two. As expected, the synthetic benchmarks did pick-up on the larger bandwidth, but only in the case of the 400 MHz FSB did I see anything approaching the theoretical increase (14 % of 25 % vs 15 % of 60 %).

If you read my first memory bandwidth post, perhaps the same conclusions can be drawn from these new data. One thing I'll add is that this new MB doesn't require extra voltage like my older P5B-Deluxe did to run the higher dividers, so it's not producing that much more heat. That said, I'm actually running the system with the 4:5 divider, since things seem to feel faster to me (windows opening, responsiveness, etc.) which are all unfortunately intangibles I can't measure.
 
Last edited:
Joined
Dec 18, 2005
Messages
8,253 (1.24/day)
System Name money pit..
Processor Intel 9900K 4.8 at 1.152 core voltage minus 0.120 offset
Motherboard Asus rog Strix Z370-F Gaming
Cooling Dark Rock TF air cooler.. Stock vga air coolers with case side fans to help cooling..
Memory 32 gb corsair vengeance 3200
Video Card(s) Palit Gaming Pro OC 2080TI
Storage 150 nvme boot drive partition.. 1T Sandisk sata.. 1T Transend sata.. 1T 970 evo nvme m 2..
Display(s) 27" Asus PG279Q ROG Swift 165Hrz Nvidia G-Sync, IPS.. 2560x1440..
Case Gigabyte mid-tower.. cheap and nothing special..
Audio Device(s) onboard sounds with stereo amp..
Power Supply EVGA 850 watt..
Mouse Logitech G700s
Keyboard Logitech K270
Software Win 10 pro..
Benchmark Scores Firestike 29500.. timepsy 14000..
i have never bothered tweaking memory over much cos its never shown me any noticeable real life gains.. only in memory bandwidth benches.. he he

having said that neither has tweaking my cpu only in cpu benches.. he he he he

trog
 
Joined
Jul 19, 2006
Messages
43,585 (6.74/day)
Processor AMD Ryzen 7 7800X3D
Motherboard ASUS TUF x670e
Cooling EK AIO 360. Phantek T30 fans.
Memory 32GB G.Skill 6000Mhz
Video Card(s) Asus RTX 4090
Storage WD m.2
Display(s) LG C2 Evo OLED 42"
Case Lian Li PC 011 Dynamic Evo
Audio Device(s) Topping E70 DAC, SMSL SP200 Headphone Amp.
Power Supply FSP Hydro Ti PRO 1000W
Mouse Razer Basilisk V3 Pro
Keyboard Tester84
Software Windows 11
Well done!:toast: Makes me feel better for running a lower FSB.
 

jonmcc33

New Member
Joined
Mar 25, 2008
Messages
580 (0.10/day)
Location
Fort Myers, FL
System Name BLACK
Processor Intel Core 2 Duo E6400 @ 3.2GHz
Motherboard Abit IP35-E @ 1600FSB
Cooling Thermalright XP-90 w/92mm ADDA fan
Memory 4GB Hynix DDR2-667 @ 800MHz
Video Card(s) PowerColor Radeon X1900XT
Storage (2) 320GB Seagate Barracuda 7200.10
Display(s) Samsung 225BW
Case Antec Nine Hundred
Audio Device(s) Audigy2 ZS Platinum
Power Supply Corsair HX520
Software Windows Vista Business x64 SP1
Yeah, similar to video card overclocking to be honest. People that overclock get XXX number of points more in 3DMark but real world the difference is not even noticeable. I flashed the BIOS of my X1900XT to an X1900XTX and noticed nothing faster at all, and some games started suffering from graphics tearing. I flashed back to the X1900XT and games were the same but no graphics tearing.
 
Joined
Nov 27, 2006
Messages
2,106 (0.33/day)
System Name Norfree
Processor i5 3570k @4.4
Motherboard Gigabyte UD5H
Cooling 212 Evo
Memory 4x4GB Kingston 1600 @ 1833 9cl
Video Card(s) Sapphire Nitro Fury
Storage Corsair SSD, WD Black
Display(s) 1080p TV
Case Corsair 300-R
Audio Device(s) Auzentech Prelude > Fidelio X2s and AD-700s
Power Supply PCP&C Silent 950w
Software Win 10 Pro 64
hmm, thats a very interesting read. it seems that game arent very memory dependent then. I guess we have a long ways to go before RAM is a bottleneck to a CPU.

What you're saying is that its pointless to buy high quality ram because it makes little difference? ddr667=ddr1066? doesnt seem right to me, but then you have the tests to prove it..unless I am reading wrong.
 
Joined
Dec 13, 2007
Messages
2,758 (0.46/day)
hmm, thats a very interesting read. it seems that game arent very memory dependent then. I guess we have a long ways to go before RAM is a bottleneck to a CPU.

What you're saying is that its pointless to buy high quality ram because it makes little difference? ddr667=ddr1066? doesnt seem right to me, but then you have the tests to prove it..unless I am reading wrong.

well not many games uses more than 2gb. Hopefully in the near future we will see more x64 base games. CS3 is one of the heaviest programs, but it more cpu then memory. But till CS4 comes out. It should balance out awhole lot more. With my specs, during any encode. I'll use 100% cpu and only 35% of memory. :ohwell:
 

jonmcc33

New Member
Joined
Mar 25, 2008
Messages
580 (0.10/day)
Location
Fort Myers, FL
System Name BLACK
Processor Intel Core 2 Duo E6400 @ 3.2GHz
Motherboard Abit IP35-E @ 1600FSB
Cooling Thermalright XP-90 w/92mm ADDA fan
Memory 4GB Hynix DDR2-667 @ 800MHz
Video Card(s) PowerColor Radeon X1900XT
Storage (2) 320GB Seagate Barracuda 7200.10
Display(s) Samsung 225BW
Case Antec Nine Hundred
Audio Device(s) Audigy2 ZS Platinum
Power Supply Corsair HX520
Software Windows Vista Business x64 SP1
hmm, thats a very interesting read. it seems that game arent very memory dependent then. I guess we have a long ways to go before RAM is a bottleneck to a CPU.

What you're saying is that its pointless to buy high quality ram because it makes little difference? ddr667=ddr1066? doesnt seem right to me, but then you have the tests to prove it..unless I am reading wrong.

Memory bandwidth dependent, no. Nothing really is. Heck, you can run your FSB higher than your memory clock and still not suffer in performance.
 

HTC

Joined
Apr 1, 2008
Messages
4,601 (0.79/day)
Location
Portugal
System Name HTC's System
Processor Ryzen 5 2600X
Motherboard Asrock Taichi X370
Cooling NH-C14, with the AM4 mounting kit
Memory G.Skill Kit 16GB DDR4 F4 - 3200 C16D - 16 GTZB
Video Card(s) Sapphire Nitro+ Radeon RX 480 OC 4 GB
Storage 1 Samsung NVMe 960 EVO 250 GB + 1 3.5" Seagate IronWolf Pro 6TB 7200RPM 256MB SATA III
Display(s) LG 27UD58
Case Fractal Design Define R6 USB-C
Audio Device(s) Onboard
Power Supply Corsair TX 850M 80+ Gold
Mouse Razer Deathadder Elite
Software Ubuntu 19.04 LTS
Since pretty much no applications use 2 GBs of memory, why not test this wile multitasking? Say ... 3 or 4 programs @ the same time to maximize memory usage, and the compare. Dunno if it would work, though.

Regarding higher speed memory, the only reason i can think of is the performance level (tRD) you can achieve with it:

http://www.anandtech.com/mb/showdoc.aspx?i=3208&p=1

Notice the gains in 3D06 by changing the tRD only! With lower speed memory, you can't get a low tRD, but you can with high speed memory.

EDIT

It's probably a good idea to include the RAM timings: 5-5-5-15 and 4-4-4-12 should have differences:



Couldn't get any lower then 5-5-5-15 with this RAM ratio



With this ratio, i was able to get 4-4-4-10.

Notice the little difference in latency and the memory read, and the very big difference on both caches.

Still, it all depends on the timings used for the original tests, IMO.
 
Last edited:
Joined
Dec 18, 2005
Messages
8,253 (1.24/day)
System Name money pit..
Processor Intel 9900K 4.8 at 1.152 core voltage minus 0.120 offset
Motherboard Asus rog Strix Z370-F Gaming
Cooling Dark Rock TF air cooler.. Stock vga air coolers with case side fans to help cooling..
Memory 32 gb corsair vengeance 3200
Video Card(s) Palit Gaming Pro OC 2080TI
Storage 150 nvme boot drive partition.. 1T Sandisk sata.. 1T Transend sata.. 1T 970 evo nvme m 2..
Display(s) 27" Asus PG279Q ROG Swift 165Hrz Nvidia G-Sync, IPS.. 2560x1440..
Case Gigabyte mid-tower.. cheap and nothing special..
Audio Device(s) onboard sounds with stereo amp..
Power Supply EVGA 850 watt..
Mouse Logitech G700s
Keyboard Logitech K270
Software Win 10 pro..
Benchmark Scores Firestike 29500.. timepsy 14000..
well not many games uses more than 2gb. Hopefully in the near future we will see more x64 base games. CS3 is one of the heaviest programs, but it more cpu then memory. But till CS4 comes out. It should balance out awhole lot more. With my specs, during any encode. I'll use 100% cpu and only 35% of memory. :ohwell:

i dont think any games use anywhere near 2 gigs of memory.. i loaded six high end-ish games all at once before i got to over 3 gig memory usage in task manager just to see.. i am useing XP thow..

trog
 

sneekypeet

Retired Super Moderator
Joined
Apr 12, 2006
Messages
29,407 (4.48/day)
System Name EVA-01
Processor Intel i7 13700K
Motherboard Asus ROG Maximus Z690 HERO EVA Edition
Cooling ASUS ROG Ryujin III 360 with Noctua Industrial Fans
Memory PAtriot Viper Elite RGB 96GB @ 6000MHz.
Video Card(s) Asus ROG Strix GeForce RTX 3090 24GB OC EVA Edition
Storage Addlink S95 M.2 PCIe GEN 4x4 2TB
Display(s) Asus ROG SWIFT OLED PG42UQ
Case Thermaltake Core P3 TG
Audio Device(s) Realtek on board > Sony Receiver > Cerwin Vegas
Power Supply be quiet DARK POWER PRO 12 1500W
Mouse ROG STRIX Impact Electro Punk
Keyboard ROG STRIX Scope TKL Electro Punk
Software Windows 11
@ greysky...I got a few questions...

#1 whos info is this, as the test specs dont match your own.

#2 so you are telling me if I run my 667 ram at 1150MHZ, I will see no difference?

#3 If this is infact your info, did you run the corsair in test with the same (way to friggin loose) subtimings.

Realistically your subtiming skills put a huge dent in the rams performance even from stock levels.
 
Joined
May 28, 2005
Messages
4,994 (0.73/day)
Location
South of England
System Name Box of Distraction
Processor Ryzen 7 1800X
Motherboard Crosshair VI Hero
Cooling Custom watercooling
Memory G.Skill TridentZ 2x8GB @ 3466MHz CL14 1T
Video Card(s) EVGA 1080Ti FE. WC'd & TDP limit increased to 360W.
Storage Samsung 960 Evo 500GB & WD Black 2TB storage drive.
Display(s) Asus ROG Swift PG278QR 27" 1440P 165hz Gsync
Case Phanteks Enthoo Pro M
Audio Device(s) Phillips Fidelio X2 headphones / basic Bose speakers
Power Supply EVGA Supernova 750W G3
Mouse Logitech G602
Keyboard Cherry MX Board 6.0 (mx red switches)
Software Win 10 & Linux Mint
Benchmark Scores https://hwbot.org/user/infrared
Fairly interesting, but you havn't mensioned subtimings or voltage anywhere.

for the 67 % increase in memory bandwidth, the marginal gains in actual performance weren't worth the extra voltage/heat.

IMO Overclocking is a compromise. It's not a choice between either stock or overclocked. Also, 1066mhz shouldn't require a lot of voltage, your ram should run @ 1066mhz with only 1.9-2.0v 5-5-5-12-[2-30-3-2-2] which would yeild good performance without getting too hot.
 
Joined
Jun 11, 2007
Messages
192 (0.03/day)
Processor i7-3770K @ 45x100
Motherboard P8Z77-V Pro
Cooling Noctua NH-D14
Memory G.SKILL Ripjaws X Series (2 x 8GB) DDR3 1600/F3-1600C9D-16GXM
Video Card(s) Onboard HD4000
Storage Vertex 4 128 GB + other HDDs
Case P183
Power Supply Seasonic SS-560KM
@ greysky...I got a few questions...

#1 whos info is this, as the test specs dont match your own.

#2 so you are telling me if I run my 667 ram at 1150MHZ, I will see no difference?

#3 If this is infact your info, did you run the corsair in test with the same (way to friggin loose) subtimings.

Realistically your subtiming skills put a huge dent in the rams performance even from stock levels.

#1 - it's mine from my system. I don't understand.
#2 - I don't understand your question.
#3 - I modified the post to answer your question as well as went from 5:6 (old post) to 4:5 (new post).
 

sneekypeet

Retired Super Moderator
Joined
Apr 12, 2006
Messages
29,407 (4.48/day)
System Name EVA-01
Processor Intel i7 13700K
Motherboard Asus ROG Maximus Z690 HERO EVA Edition
Cooling ASUS ROG Ryujin III 360 with Noctua Industrial Fans
Memory PAtriot Viper Elite RGB 96GB @ 6000MHz.
Video Card(s) Asus ROG Strix GeForce RTX 3090 24GB OC EVA Edition
Storage Addlink S95 M.2 PCIe GEN 4x4 2TB
Display(s) Asus ROG SWIFT OLED PG42UQ
Case Thermaltake Core P3 TG
Audio Device(s) Realtek on board > Sony Receiver > Cerwin Vegas
Power Supply be quiet DARK POWER PRO 12 1500W
Mouse ROG STRIX Impact Electro Punk
Keyboard ROG STRIX Scope TKL Electro Punk
Software Windows 11
#1 your system spec's<<<<<< dont come close to the testing equiptment.

#2 you are saying in very simplified terms that there arent gains worth enough to OC my ram. So with my set of 667MHz ram running at 1150MHz I will see no real world performance gains?

#3 by leaving all subtimings on auto then?
 
Joined
Jun 11, 2007
Messages
192 (0.03/day)
Processor i7-3770K @ 45x100
Motherboard P8Z77-V Pro
Cooling Noctua NH-D14
Memory G.SKILL Ripjaws X Series (2 x 8GB) DDR3 1600/F3-1600C9D-16GXM
Video Card(s) Onboard HD4000
Storage Vertex 4 128 GB + other HDDs
Case P183
Power Supply Seasonic SS-560KM
I edited the first post switching the highest 400 MHz FSB run from 5:6 to 4:5 (960 MHz vs. 1,000 MHz) and included some info about subtimings to make things more clear.
 

HTC

Joined
Apr 1, 2008
Messages
4,601 (0.79/day)
Location
Portugal
System Name HTC's System
Processor Ryzen 5 2600X
Motherboard Asrock Taichi X370
Cooling NH-C14, with the AM4 mounting kit
Memory G.Skill Kit 16GB DDR4 F4 - 3200 C16D - 16 GTZB
Video Card(s) Sapphire Nitro+ Radeon RX 480 OC 4 GB
Storage 1 Samsung NVMe 960 EVO 250 GB + 1 3.5" Seagate IronWolf Pro 6TB 7200RPM 256MB SATA III
Display(s) LG 27UD58
Case Fractal Design Define R6 USB-C
Audio Device(s) Onboard
Power Supply Corsair TX 850M 80+ Gold
Mouse Razer Deathadder Elite
Software Ubuntu 19.04 LTS
#2 you are saying in very simplified terms that there arent gains worth enough to OC my ram. So with my set of 667MHz ram running at 1150MHz I will see no real world performance gains?

In my case, i went the other way around:

I've lowered from 1062 to 884 and it's faster, @ least in Nero Recode it is.
 
Top