x264 video encoding benchmark

graysky · Post by **graysky** » Sat Sep 08, 2007 5:42 pm

I put together a self-contained x264 video encoding benchmark. Techarp kindly agreed to host the file and results at this URL.

Basically, you run the test encode and it will report back frames-per-second values for your machine @ it's clock/overclock level. You can run it at your stock settings and at your overclock settings to see how your machine compares to others in the database.

The database is small right now (as of 08-sep), but as you guys report in results, I will populate it. My goal is to have a representative set of data for many different chips and chipsets. Hopefully, we'll get some Penryn and Phenom data when they become available. Also, if anyone out here has some of the high end AMD chips, please contribute. Instructions and the file are at that url.

Also, please report your results here in this thread. I will keep the data at that url to keep things simple.

Thanks all.

SebRad · Post by **SebRad** » Sun Sep 09, 2007 12:40 am

Hi, well I've just run the test on my machine clocked as I normally use it. No clean boot or anything, just shut everything down and ran the test.

Core 2 Duo E6600 (2.4GHz, 4MB) @ 3326 (475 x 7), 2x1GB DDR2 950 5,5,5,15 dual channel, Asus P5B-E Plus P965 motherboard. (WD 5000AAKS, ATi X1950pro, etc etc) Vista Home Premium 32bit
Results
---------- RUN1PASS1.LOG
encoded 1749 frames, 92.07 fps, 1850.89 kb/s

---------- RUN2PASS1.LOG
encoded 1749 frames, 91.79 fps, 1850.89 kb/s

---------- RUN3PASS1.LOG
encoded 1749 frames, 92.63 fps, 1850.89 kb/s

---------- RUN4PASS1.LOG
encoded 1749 frames, 92.76 fps, 1850.89 kb/s

---------- RUN5PASS1.LOG
encoded 1749 frames, 92.49 fps, 1850.89 kb/s

---------- RUN1PASS2.LOG
encoded 1749 frames, 22.89 fps, 1826.37 kb/s

---------- RUN2PASS2.LOG
encoded 1749 frames, 22.94 fps, 1826.38 kb/s

---------- RUN3PASS2.LOG
encoded 1749 frames, 22.95 fps, 1826.38 kb/s

---------- RUN4PASS2.LOG
encoded 1749 frames, 22.75 fps, 1826.33 kb/s

---------- RUN5PASS2.LOG
encoded 1749 frames, 22.99 fps, 1826.38 kb/s
Will edit this and add other clockspeeds eventually. Regards, Seb

graysky · Post by **graysky** » Sun Sep 09, 2007 2:20 am

SebRad wrote:Core 2 Duo E6600 (2.4GHz, 4MB) @ 3326 (475 x 7), 2x1GB DDR2 950 5,5,5,15 dual channel, Asus P5B-E Plus P965 motherboard. (WD 5000AAKS, ATi X1950pro, etc etc) Vista Home Premium 32bit

Thanks for the result. What is your mem clocked @ (like 475 MHz or ...?)

Also, would you be willing to run the same benchmark @ stock settings for comparison (9x266)?

ACook · Post by **ACook** » Sun Sep 09, 2007 5:22 pm

If you don't have AVISynth 2.5.7 on your machine, you'll need to install it before running the benchmark.

what happens if you have a diff version of avs installed? and when I already have a version of dgindex.dll in my avs plugins? no id which version.

graysky · Post by **graysky** » Mon Sep 10, 2007 12:21 pm

ACook wrote:
If you don't have AVISynth 2.5.7 on your machine, you'll need to install it before running the benchmark.
what happens if you have a diff version of avs installed? and when I already have a version of dgindex.dll in my avs plugins? no id which version.

I believe if the version you have is older, the avs will not work. What you can do is temporarly rename your current dgindex.dll to dgindex.current and copy the one from the benchmark to the /avisynth/plugins to run the benchmark. When you're finished, simply delete the dgindex.dll from the benchmark and rename your original one back.

ddrueding1 · Post by **ddrueding1** » Mon Sep 10, 2007 4:04 pm

A64 X2 6000+ Stock Speed, 4x1GB DDR2 800 5-5-5-15, see sig for more details.

Results:
---------- RUN1PASS1.LOG
encoded 1749 frames, 75.25 fps, 1850.89 kb/s

---------- RUN2PASS1.LOG
encoded 1749 frames, 75.21 fps, 1850.89 kb/s

---------- RUN3PASS1.LOG
encoded 1749 frames, 75.57 fps, 1850.89 kb/s

---------- RUN4PASS1.LOG
encoded 1749 frames, 75.50 fps, 1850.89 kb/s

---------- RUN5PASS1.LOG
encoded 1749 frames, 75.22 fps, 1850.89 kb/s

---------- RUN1PASS2.LOG
encoded 1749 frames, 18.58 fps, 1826.37 kb/s

---------- RUN2PASS2.LOG
encoded 1749 frames, 18.58 fps, 1826.38 kb/s

---------- RUN3PASS2.LOG
encoded 1749 frames, 18.63 fps, 1826.37 kb/s

---------- RUN4PASS2.LOG
encoded 1749 frames, 18.70 fps, 1826.38 kb/s

---------- RUN5PASS2.LOG
encoded 1749 frames, 18.67 fps, 1826.38 kb/s

ACook · Post by **ACook** » Tue Sep 11, 2007 7:00 am

k, my lowly A64 3000+ winchester, 9x200,9 according to cpuz. so not oc'd.

also, when running this, I had 10 tabs open on opera including irc, and some other programs idling in the bg, while seeding stuff on azureus. can't imagine that having much effect on overall performance though. for the duration of the test I did not increase the speed on the zalman from 25%, and temp didn't go over 51C, which was nice. idle temp is 40. thinking of upgrading to x2 3800 to squeeze another year out of this thing, wonder what that will do to my temps.

edit: full specs:
AMD Athlon 64 3000+ Winchester Socket939, 9 x 200.9MHz, NForce4
mem: 3-3-3-8 1T @ 200.9MHz

think that's all stock.

---------- RUN1PASS1.LOG
encoded 1749 frames, 23.52 fps, 1854.10 kb/s

---------- RUN2PASS1.LOG
encoded 1749 frames, 23.12 fps, 1854.10 kb/s

---------- RUN3PASS1.LOG
encoded 1749 frames, 23.88 fps, 1854.10 kb/s

---------- RUN4PASS1.LOG
encoded 1749 frames, 23.49 fps, 1854.10 kb/s

---------- RUN5PASS1.LOG
encoded 1749 frames, 23.54 fps, 1854.10 kb/s

---------- RUN1PASS2.LOG
encoded 1749 frames, 5.36 fps, 1825.89 kb/s

---------- RUN2PASS2.LOG
encoded 1749 frames, 5.43 fps, 1825.89 kb/s

---------- RUN3PASS2.LOG
encoded 1749 frames, 5.48 fps, 1825.89 kb/s

---------- RUN4PASS2.LOG
encoded 1749 frames, 5.55 fps, 1825.89 kb/s

---------- RUN5PASS2.LOG
encoded 1749 frames, 5.32 fps, 1825.89 kb/s

ps: avs 2.5.6 was installed, didn't change.

protellect · Post by **protellect** » Tue Sep 11, 2007 10:14 am

AMD 5600X2, Crucial DDR2-800 4-4-4-12. Nvidia 590 Chipset, Asus board, stock timings.

Seems kinda slow

---------- RUN1PASS1.LOG
encoded 1749 frames, 74.32 fps, 1850.89 kb/s

---------- RUN2PASS1.LOG
encoded 1749 frames, 74.52 fps, 1850.89 kb/s

---------- RUN3PASS1.LOG
encoded 1749 frames, 74.48 fps, 1850.89 kb/s

---------- RUN4PASS1.LOG
encoded 1749 frames, 74.33 fps, 1850.89 kb/s

---------- RUN5PASS1.LOG
encoded 1749 frames, 74.43 fps, 1850.89 kb/s

---------- RUN1PASS2.LOG
encoded 1749 frames, 18.20 fps, 1826.37 kb/s

---------- RUN2PASS2.LOG
encoded 1749 frames, 18.23 fps, 1826.38 kb/s

---------- RUN3PASS2.LOG
encoded 1749 frames, 18.14 fps, 1826.38 kb/s

---------- RUN4PASS2.LOG
encoded 1749 frames, 18.18 fps, 1826.37 kb/s

---------- RUN5PASS2.LOG
encoded 1749 frames, 18.17 fps, 1826.37 kb/s

edit; 4-3-3-10 memory timings

---------- RUN1PASS1.LOG
encoded 1749 frames, 75.17 fps, 1850.89 kb/s

---------- RUN2PASS1.LOG
encoded 1749 frames, 75.33 fps, 1850.89 kb/s

---------- RUN3PASS1.LOG
encoded 1749 frames, 75.28 fps, 1850.89 kb/s

---------- RUN4PASS1.LOG
encoded 1749 frames, 75.23 fps, 1850.89 kb/s

---------- RUN5PASS1.LOG
encoded 1749 frames, 74.77 fps, 1850.89 kb/s

---------- RUN1PASS2.LOG
encoded 1749 frames, 18.35 fps, 1826.38 kb/s

---------- RUN2PASS2.LOG
encoded 1749 frames, 18.40 fps, 1826.38 kb/s

---------- RUN3PASS2.LOG
encoded 1749 frames, 18.40 fps, 1826.37 kb/s

---------- RUN4PASS2.LOG
encoded 1749 frames, 18.40 fps, 1826.37 kb/s

---------- RUN5PASS2.LOG
encoded 1749 frames, 18.43 fps, 1826.37 kb/s

graysky · Post by **graysky** » Tue Sep 11, 2007 12:09 pm

@acook and protellect - thanks for the results, guys... also what O/S are you each running?

ACook · Post by **ACook** » Tue Sep 11, 2007 3:49 pm

xp sp2 32bit

hope someone posts a stock X2 3800 or 4200 soon, want to know if it's worth the price diff

protellect · Post by **protellect** » Tue Sep 11, 2007 4:48 pm

same, xp 32 bit sp2

graysky · Post by **graysky** » Tue Sep 11, 2007 11:11 pm

cool, thanks guys

ceselb · Post by **ceselb** » Sat Sep 15, 2007 8:37 am

Intel Pentium 4 Williamette 1.5Ghz, 15x100, 768MB dual channel RDRAM 400Mhz Timings 11,9. Intel D850GB i850, XP Pro SP2.

---------- RUN1PASS1.LOG
encoded 1749 frames, 9.28 fps, 1854.10 kb/s

---------- RUN2PASS1.LOG
encoded 1749 frames, 9.39 fps, 1854.10 kb/s

---------- RUN3PASS1.LOG
encoded 1749 frames, 9.21 fps, 1854.10 kb/s

---------- RUN4PASS1.LOG
encoded 1749 frames, 9.37 fps, 1854.10 kb/s

---------- RUN5PASS1.LOG
encoded 1749 frames, 9.40 fps, 1854.10 kb/s

---------- RUN1PASS2.LOG
encoded 1749 frames, 2.11 fps, 1825.89 kb/s

---------- RUN2PASS2.LOG
encoded 1749 frames, 2.12 fps, 1825.89 kb/s

---------- RUN3PASS2.LOG
encoded 1749 frames, 2.02 fps, 1825.89 kb/s

---------- RUN4PASS2.LOG
encoded 1749 frames, 2.01 fps, 1825.89 kb/s

---------- RUN5PASS2.LOG
encoded 1749 frames, 2.12 fps, 1825.89 kb/s

PS: Yes, this is my main box. Yes, it belongs in a museum. Yes, I'm going to upgdrade soon.

BillyBuerger · Post by **BillyBuerger** » Sat Sep 15, 2007 9:34 am

I was really tempted to try this on a new server we just got at work with two Xeon X5355 (8 cores total). If I would have seen this a week earlier I probably could have. I did install Windows Server 2003 on it at first just to get it up and running. But now I've got VMWare and a couple VMs running on it. I can't afford to take it down just for something fun. I did run it on a VM with 4 virtual CPUs. But the results were right at the bottom of the Quad-core CPUs. The Fully Buffered memory and virtual machine overhead probably don't help.

graysky · Post by **graysky** » Sat Sep 15, 2007 11:34 am

@cescelb - thanks for the result
@Billy - I'm sure you can run the benchmark during some 'scheduled downtime'

In all seriousness, it would be pretty cool to see how your octa core machine scales x264.

notquitequiet · Post by **notquitequiet** » Sat Sep 15, 2007 5:02 pm

...but, the program kept running.

X2 4400, 2 GB DDR400 dual channel, 6150 chipset and XP 32 bit
stock clocked

---------- RUN1PASS1.LOG
encoded 1749 frames, 58.79 fps, 1850.89 kb/s

---------- RUN2PASS1.LOG
encoded 1749 frames, 58.51 fps, 1850.89 kb/s

---------- RUN3PASS1.LOG
encoded 1749 frames, 58.76 fps, 1850.89 kb/s

---------- RUN4PASS1.LOG
encoded 1749 frames, 59.16 fps, 1850.89 kb/s

---------- RUN5PASS1.LOG
encoded 1749 frames, 58.91 fps, 1850.89 kb/s

---------- RUN1PASS2.LOG
encoded 1749 frames, 14.01 fps, 1826.38 kb/s

---------- RUN2PASS2.LOG
encoded 1749 frames, 14.03 fps, 1826.37 kb/s

---------- RUN3PASS2.LOG
encoded 1749 frames, 14.11 fps, 1826.37 kb/s

---------- RUN4PASS2.LOG
encoded 1749 frames, 14.15 fps, 1826.37 kb/s

---------- RUN5PASS2.LOG
encoded 1749 frames, 14.15 fps, 1826.37 kb/s

I did some other things while it was running, but not much. Like I said, I fell asleep.

Max Slowik · Post by **Max Slowik** » Sat Sep 15, 2007 6:19 pm

Cool utility, I'll see if I can add it to our motherboard reviews.

Oh, right, I'm in charge of motherboard reviews.

graysky · Post by **graysky** » Sun Sep 16, 2007 1:26 am

notquitequiet wrote:...but, the program kept running.

X2 4400, 2 GB DDR400 dual channel, 6150 chipset and XP 32 bit
stock clocked

Cool man, thanks for the results. Can you report your mem timings and final mem core rate (i.e. 5-5-5-15 @ 400 MHz) Also who makes your motherboard? I never heard of a 6150 chipset.

Thanks!

Mikael · Post by **Mikael** » Sun Sep 16, 2007 2:48 am

Pentium Dual Core E2160 @ 3.00GHz (9*333)
4GB DDR2-833MHz 5-5-5-15
Asus P5B
Windows XP SP2

---------- RUN1PASS1.LOG
encoded 1749 frames, 81.76 fps, 1850.89 kb/s

---------- RUN2PASS1.LOG
encoded 1749 frames, 81.89 fps, 1850.89 kb/s

---------- RUN3PASS1.LOG
encoded 1749 frames, 81.88 fps, 1850.89 kb/s

---------- RUN4PASS1.LOG
encoded 1749 frames, 81.82 fps, 1850.89 kb/s

---------- RUN5PASS1.LOG
encoded 1749 frames, 82.00 fps, 1850.89 kb/s

---------- RUN1PASS2.LOG
encoded 1749 frames, 19.44 fps, 1826.21 kb/s

---------- RUN2PASS2.LOG
encoded 1749 frames, 19.47 fps, 1826.37 kb/s

---------- RUN3PASS2.LOG
encoded 1749 frames, 19.44 fps, 1826.38 kb/s

---------- RUN4PASS2.LOG
encoded 1749 frames, 19.46 fps, 1826.38 kb/s

---------- RUN5PASS2.LOG
encoded 1749 frames, 19.44 fps, 1826.38 kb/s

I'm kind of proud of my super cheap little CPU!

It's faster clock-for-clock than the X2 6000+ and probably consumes atleast 40W less.

graysky · Post by **graysky** » Sun Sep 16, 2007 4:35 am

@Mikael - Jag mÃ¥r bra, tack!

Mikael · Post by **Mikael** » Sun Sep 16, 2007 5:20 am

graysky wrote:@Mikael - Jag mÃ¥r bra, tack!

Ã…hh, Ã¤r du svensk? Och varfÃ¶r skrev du "Jag mÃ¥r bra, tack!"?

floffe · Post by **floffe** » Sun Sep 16, 2007 6:35 am

graysky wrote:Cool man, thanks for the results. Can you report your mem timings and final mem core rate (i.e. 5-5-5-15 @ 400 MHz) Also who makes your motherboard? I never heard of a 6150 chipset.

There are several manufacturers making mobos with nForce 6150 chipsets with integrated graphics, pretty much all of them mATX. Hmmm, looking around they seem to be shifted out a bit, probably in favour of the newer nvidia integrated graphics chipset. Newegg still has Foxconn, Asus, DFI and Abit versions of it, but I know that Gigabyte and MSI also had mobos based on it out.

graysky · Post by **graysky** » Sun Sep 16, 2007 7:52 am

Mikael wrote:
graysky wrote:@Mikael - Jag mÃ¥r bra, tack!
Ã…hh, Ã¤r du svensk? Och varfÃ¶r skrev du "Jag mÃ¥r bra, tack!"?

Sorry dude, I have no idea how to speak the language. I just looked up Swedish on wikipedia and there was a list of common phrase

jaganath · Post by **jaganath** » Sun Sep 16, 2007 9:10 am

Mikael wrote:
graysky wrote:@Mikael - Jag mÃ¥r bra, tack!
Ã…hh, Ã¤r du svensk? Och varfÃ¶r skrev du "Jag mÃ¥r bra, tack!"?

I have minimal knowledge of swedish, but i think he said "Ah, are you swedish? and why did you write "I am fine, thanks!"? do I get a cookie?

floffe · Post by **floffe** » Sun Sep 16, 2007 9:58 am

jaganath wrote: do I get a cookie?

You sure do! (Yes, I'm Swedish too)

notquitequiet · Post by **notquitequiet** » Sun Sep 16, 2007 10:16 am

graysky wrote: Cool man, thanks for the results. Can you report your mem timings and final mem core rate (i.e. 5-5-5-15 @ 400 MHz) Also who makes your motherboard? I never heard of a 6150 chipset.

Sure.

nVidia made the chipset. Similar to the 6100 and, from what I've read, the 7025 and 7050 chipsets.

It's Kingston Value RAM running at SPD DDR-400 PC3200. Nothing exotic. No heatspreaders or RAM fans. I looked in the BIOS, but there's nothing but "AUTO" in the settings. SPD.

It's a MSI K8NGM2-FID motherboard. I don't know if this long URL will work. http://global.msi.com.tw/index.php?func ... =&cat3_no=
Nothing fancy about it.

CPU-Z has the RAM timings. 3-3-3-8 @ 200MHz. CPU is 11 x 200 MHz. Code name Toledo.

ACook · Post by **ACook** » Sun Sep 16, 2007 2:19 pm

thanks notquitequiet for that data, I've just ordered the 4200+ for my s939 board, we'll see how it compares to the 4400+.
looks I'll double my speed at least, for x264. hope it also impacts xvid encoding with agk like that.

perhaps ill do some overclocks first on the old 3000+ just to see how fast it can go

ceselb · Post by **ceselb** » Sun Sep 16, 2007 2:25 pm

floffe wrote:Yes, I'm Swedish too

Me three! Even the same city, cool.

Mikael · Post by **Mikael** » Sun Sep 16, 2007 3:23 pm

Forgot to mention earlier: I'll be re-running the test at the exact same frequencies but using my E6600 instead (4MB L2 instead of 1MB).

graysky wrote:Sorry dude, I have no idea how to speak the language. I just looked up Swedish on wikipedia and there was a list of common phrase

Heheh... Thought so. Always nice to see some swedish on predominantly english boards, though.

jaganath wrote:I have minimal knowledge of swedish, but i think he said "Ah, are you swedish? and why did you write "I am fine, thanks!"? do I get a cookie?

Impressive! You got it just right. *hands over giant cookie*

graysky · Post by **graysky** » Thu Sep 20, 2007 1:41 pm

As of 20-Sep-2007, we have data on over 100 Intel-based systems and on over 40 AMD-based systems. There are a few trends I picked-up on while browsing through the database. I put them into a single table and color coded them to make them easier to see. If you see a trend I missed, lemme know and I'll add it to the table.

Request: we don't have a single example of a machine that has both WinXP and WinVista on it. If you have a dual-boot setup, it would be cool to see the difference the O/S makes. Another missing trend is a 32-bit O/S vs. the same O/S that's 64-bit.

On to the table:

Yellow: Nearly 1:1 increase by adding an additional processor to a dual-chip MB
Orange: Some operating systems seem to handle x264 more efficiently than others
Red: Insignificant gain by upping the DRAM speed by 50 %
Blue: For the most part, these chips scale in a pretty linear fashion
Green: Tighter/looser memory timings have a pretty insignificant effect
Purple: Keeping the same over-all clock speed using a different combo of multiplier and FSB can give pretty insignificant gains

Again, I only gave this a once-over look; please point out any trends you see that I missed and also don't forgot about the O/S request!

Thanks again to all who contributed!

x264 video encoding benchmark

x264 video encoding benchmark

I fell asleep

Re: I fell asleep

Re: I fell asleep

Re: I fell asleep