Multiple Encoder Processes: My experience

[rpboy]

So I thought that I would run a test to see how much faster selecting the "Multiple Encoder Processes" option would be on a pretty simple DVD.

My computer: Pentium 4 3.0 GHz HyperThreaded-enabled processor with 800 MHz FSB, 1 GB RAM, Windows XP SP2.

I tested using DVD-Rebuilder 1.07.1 Pro version and HCEnc 0.17 (in Best mode, using Auto-Q2 matrix).

The DVD had just one VTS to be processed. As the menu is contained in the VIDEO_TS.VOB file, no menu encoding was done. Size of original disc was 5.17 GB.

With the "Multiple Encoder Processes" option deselected, Windows Task Manager reports 50-54% CPU usage while encoding with HCEnc, so it seemed like a sizable benefit could be optained with the option selected.

I ran the same DVD through with the "Multiple Encoder Processes" option selected and then deselected.

Final results
"Multiple Encoder Processes" option selected: 184 minutes

"Multiple Encoder Processes" option deselected: 203 minutes

Kind of surprising. It wasn't a huge difference, about a 10% speedup. I thought it would be better than that. I wonder if only having 1 GB of memory limited it in any way.

Anyway, would love to hear what reductions in overall time others are getting, both with HT-enabled processors as well as dual-core processors. Might convince me to cave in and go buy that Athlon 64 X2 CPU I've been ogling. ;)

[urvieh]

Hyperthreading doesn't mean double power :-). It is just a minor improvement afa calculation speed is concerned.

[rpboy]

Hyperthreading doesn't mean double power :-). It is just a minor improvement afa calculation speed is concerned.
I didn't think it would reduce the time by half. But being that the CPU is only being used at 50% in a single encoder process, I thought there would be a little more of a substantial boost than 10%.

[jdobbs]

But it's still using the same disc and memory. The effect of multiple processes will vary tremendously from computer to computer. I have other reports that show improvements of over 50% by some -- and none on others.

[Sophoclesdrf]

When a processor is showing only 50% it doesn't mean that it is more efficient, it only means that for whatever reason it is using less. An equally fast or faster machine that shows 90% CPU processes is the one that's hard at work.

[markrb]

10% is still a big gain.
It's still 1 CPU afterall, just altered to act in some respects as more then 1.
However before you jump the gun and get an X2 consider these things.

In this type of application Intel has the edge over Amd.
Like jdobbs said you still are dealing with one data bus.

I am a huge Amd fan myself and have a X2 3800.
Personally I won't go Intel right now, but if I was strictly doing video or audio encoding I would.

In the end if you decide to upgrade consider what you are doing and choose accordingly.

Mark

[Sophoclesdrf]

markrb

I already have a Dual Core, and no Intel is not faster than the right X2 even with Hyper Threading. I've considered the possibility that a dual core Intel with hyper threading could make use of the dual mode running two multi threaded apps at the same time since Hyper Threading would allow for up to 4 threads running at once. I thought that might yield some benefits in comparison to an Intel running only two threads at once on a single instance, because hyper threading would make use of the unused CPU cycles. I never considered for a moment however that it would result in faster than AMD processing only faster than one instance of an application on an Intel DC processor.

I have an Opteron 175 2.2 Ghz@2.67 Ghz and (2 x 1GB) TWINX2048-3500LLPRO on an Asus A8N-SLI Deluxe and I don't believe that the current fastest DC Intel even running Dual RB can match my 175's encoding speeds.:) And my processor generally is using better than 90% of its CPU cycles.

[gilligan2]

markrb

I already have a Dual Core, and no Intel is not faster than the right X2 even with Hyper Threading. I've considered the possibility that a dual core Intel with hyper threading could make use of the dual mode running two multi threaded apps at the same time since Hyper Threading would allow for up to 4 threads running at once. I thought that might yield some benefits in comparison to an Intel running only two threads at once on a single instance, because hyper threading would make use of the unused CPU cycles. I never considered for a moment however that it would result in faster than AMD processing only faster than one instance of an application on an Intel DC processor.

I have an Opteron 175 2.2 Ghz@2.67 Ghz and (2 x 1GB) TWINX2048-3500LLPRO on an Asus A8N-SLI Deluxe and I don't believe that the current fastest DC Intel even running Dual RB can match my 175's encoding speeds.:) And my processor generally is using better than 90% of its CPU cycles.

Yes,i can vouch for this also. I have a dual Xeon 3.06 GHz machine with 2 GB's of ram,SATA raid,etc., and also have a dual core Opteron 165 overclocked to 2.73 GHz with a DFI Lanparty,2 GB's of ram,SATA II raid 0 set up and the Opteron is right around 33% faster doing identical projects.
So even though i am losing 20% pure clockspeed (3 Ghz to 2.7 Ghz),my encodes are 33% faster using the AMD set up.
I have one of both so i am not a basher of either one but i think for them to run equal i would need a dual Xeon or dual core at around 4 GHz or possibly higher and with the reports i've seen on the heat generated by the Intel systems I will wait for a while to upgrade my second machine.

All i say is thank goodness we have at least two companies pushing each other for technology LOL

g:):D

[Sophoclesdrf]

All i say is thank goodness we have at least two companies pushing each other for technology LOL

I second that. I still have my old 2.8Ghz Northwood@3.149 Ghz and I'm not about to get rid of it. In fact I often use it for encoding because it's generally not needed.

[nwg]

I am hoping to get either a X2 4200 or X2 4600. When using this option, will each core encode a single m2v at the same time?

Will it use both cores at the same time to encode one m2v much quicker?

I use Procoder mainly, but sometimes CCE and HCEncoder as well.

[rahzel]

sorry, not much help, but you might want to know this.

i just bought a 3800+ X2, and i was able to overclock it to 2.7Ghz with no bump in voltage, and dual prime95 stable. i am using an aftermarket cooler (arctic freezer 64 pro) but at 2.7ghz, this should be faster than even a 4800+ X2, and similar, if not better than even a 5000+ X2.

to match a 4600+ x2 at stock, all you would need is to reach 2.4ghz, which should be easily attainable with a 3800+ X2. i got my 3800+ x2 and arctic freezer 64 pro cooler, all for 210 shipped.

of course it depends on the stepping, but i'm sure the steppings are currently the same pretty much anywhere.

just thought you might like to know, if you want to save some cash.

[nwg]

I have never risked overclocking as I saw my friend destroy his processor by overclocking too much.

How big is the arctic freezer 64 pro cooler compared to the AMD heatsink? I don't have much vertial room in my case.

I can get a X2 4200 for £125 and a X2 4600 for £155.

[rahzel]

overclocking is safe, as long as you know what you're doing.

what you want to do (with a64's) is keep your temperatures below 55c load, raise the voltage no higher than 1.6v (unless you have water cooling) and test for stability. theres also other things you need to know, so i guess if you're not comfortable with overclocking, i'd recommend not doing it.

the freezer 64 is pretty big overall, but small at the base. there are some fins (aluminum) that might get in the way, but they can be bent.

i dont know about the prices there, but the 3800+ is generally around 180cad, the 4200+ is about 220cad, and the 4600+ is about 280cad.

btw, like i said before, all you would need is to reach 2.4ghz to match a 4600+ at stock; the stock cooler would probably be enough to reach this.

[blutach]

Interesting info rahzel. I have a 3800 X2 not overclocking (ie 2000Mhz). Will see what can be done here.

Thanks!

Regards

[wmansir]

For anyone running CCE it is vital that your system be 100% stable. The program is THE most sensitive to overclocking/instability that I've ever encountered. I've had systems overclocked that could run prime95 and memtest for days but CCE will crash within an hour.

[nwg]

For anyone running CCE it is vital that your system be 100% stable. The program is THE most sensitive to overclocking/instability that I've ever encountered. I've had systems overclocked that could run prime95 and memtest for days but CCE will crash within an hour.


I have heard about that as well. I have no intention to overclock. I rather pay a little more knowing it is stable from the beginning.

[jdobbs]

I used to be an avid overclocker... but at some point you eventually catch yourself blaming other pieces of hardware/software for strange quirks that almost never happen. Then when you set your clock back to standard they all magically go away.

Go back through the threads here and it's amazing how many overclocking instabilities CCE has caught.

[rahzel]

i ALWAYS overclock, and i've never had a problem. never had my computer crash/freezer, never got an error in a cpu intensive program (after running prime95 and making sure my system is stable that is), never fried a CPU because i went too far etc.

but again, if you are not comfortable with it, and/or you don't know what you're doing, i recommend not doing it, but you get SO much more value out of your hardware by doing so. like i got my 3800+ X2 that would surpass a 4800+ X2 for around half the price, this is why i always buy the slowest processor, and just overclock it.

[BigDid]

...I just bought a 3800+ X2, and i was able to overclock it to 2.7Ghz with no bump in voltage, and dual prime95 stable. i am using an aftermarket cooler (arctic freezer 64 pro) but at 2.7ghz, this should be faster than even a 4800+ X2, and similar, if not better than even a 5000+ X2.

to match a 4600+ x2 at stock, all you would need is to reach 2.4ghz, which should be easily attainable with a 3800+ X2. i got my 3800+ x2 and arctic freezer 64 pro cooler, all for 210 shipped.
Hi,

Interesting. There are quite a few threads related to dual-cores/AMD/new prices etc...

I am pondering buying a 3800x2 or 4200x2 wondering if the 4200 would reach 4600 and you just answered the 3800x2 can reach even 3700:eek: . I searched on the overclokers forum and there are a lot of high overclocks with the 3800x2 rather than 4200x2. But, now than they are cheaper I'll wait a little more to see if more positives testimonies arrives.

Regarding the concept and use of overclocking, actual Mobos usually have a semi-auto mode for noob-safe/proof mode.
For the rock solid encoding mode, there are solutions like the Right mark cpu clock utility which let you configure a lot of params.
My S754 A64-2000 is overclocked to 2460Mhz with 1.5V (stock) uses the Right mark cool and quiet and throttle down to 980 (x4) with 0.95V; this for usual tasks.
When encoding (AGK/xvid with VirtualDubMod) I use another profile with a unique x9.5 multiplier(2333), 1.4v that rarely goes above 55°c, rock solid.

Any/other x2 A64 AMD experiences, overcloked or not welcome.

Did

[rahzel]

any AMD X2 processor generally should reach similar numbers, whether it be a 3800+, 4200+, 4600+ etc. whats more important is the stepping on the processor - all processors have a stepping, and the stepping is like a code, which usually has the date the processor was made, along with some letters. usually, the batches of CPU's that are made in the same time period, generally overclock very similarly, and some batches overclock better than others. so don't expect to hit 2.7ghz like i did, but again, you should be able to reach at least 2.4ghz fairly easily with any 3800+ X2.

[nwg]

I have seen a test where the X2 4200 was overclocked to 2.7ghz.

[BigDid]

... so don't expect to hit 2.7ghz like i did, but again, you should be able to reach at least 2.4ghz fairly easily with any 3800+ X2.
I have seen a test where the X2 4200 was overclocked to 2.7ghz.
Thanks for inputs. I'd rather go for a 4200x2 if 2600Mhz is reachable cause with a 3800 @ 2400 I will have no improvement on my actual processor for mono-threaded apps.

Did

[rahzel]

Thanks for inputs. I'd rather go for a 4200x2 if 2600Mhz is reachable cause with a 3800 @ 2400 I will have no improvement on my actual processor for mono-threaded apps.

Did
like i said before, ALL X2 AMD processors should reach similar numbers, meaning all X2 AMD processors will generally hit the same limit. the only advantage the higher AMD X2 processors have, is the higher multiplier.

at stock, this is how they all run:
3800+ X2 - 200x10=2.0Ghz
4200+ X2 - 200x11=2.2Ghz
4600+ X2 - 200x12=2.4Ghz

but just because the 4600+ X2 is already faster at stock, doesnt mean it will reach a higher overclock.

[BigDid]

Hi,

I made a survey of my usual online store with the customers feedback (over 1 year covered), here are the results:

AMD 64 x2 4200
12 buy, 4 overclocks
2@ 2.6
1@ 2.66
1@ 2.7

AMD 64 x2 3800
22 buy, 7 overclocks
1@ 2.1
1@ 2.2
3@ 2.5
2@ 2.6

Did

[SiKKo]

rahzel is exactly right, there really is no point buying the more expensive 4200+ x2 when the 3800+ x2 should overclock to the same speeds. Of course, a lot of the overclockability is based on the motherboard and RAM as well. I recommend DFI boards myself, as they will tend to allow you to overclock much higher than a cheap mobo. Quality RAM is also imporant. If you are interested in overclocking AMD chips, check out this forum at http://www.dfi-street.com/forum/forumdisplay.php?f=9

There is also a thread on there specifically referencing different 3800+ steppings, where people are hitting 2.8-3.0ghz. For a $150 chip, that is very tough to beat!

http://www.dfi-street.com/forum/showthread.php?t=61749

Myself, I have an Opteron 165 dual core running at 2.8ghz (air cooled @ 1.45V) and 2GB of RAM, and a 4 pass CCE encode with DVD-RB takes roughly 3 hours for an average 7.5GB movie.

[shoarthing]

rahzel is exactly right, there really is no point buying the more expensive 4200+ x2 when the 3800+ x2 should overclock to the same speeds. Of course, a lot of the overclockability is based on the motherboard and RAM as well . . . . . . . well, yes, there's not much point in praising the cheapness of the CPU if you have to use expensive RAM to reach the fairly high HTs [260~270MHz] which these claimed overclocks require . . . especially since you need a fair bit of the stuff [around 1.2GB free], if, for instance, running a couple of instances of Procoder 2.x.

I've seen too many folks trying to track down/debug 0x9c errors with AMD64 CPUs to recommend 'clocking 'em for MPEG2 re-encoding, which, as stated above, needs a genuinely solid system.

I use dual~Socket940 [ie Opteron 2xx] workstations with NUMA memory configurations, which thanks to DVDRB's support of multiple encoder instances can now exploit Procoder's mastering quality within a tolerable timeframe for the whole process - again, around 3 hours or a bit less for a typical backup.

[rahzel]

memory bandwidth/memory timings play a fairly minimal part in performance with a64's, so expensive, good overclocking ram is not a necessity for overclocking a64's, thanks to memory dividers. all motherboards have memory dividers, and overclocking friendly boards, like DFI, have many options so you can run your memory slower, but still raise the HTT.

i finally got a chance to try out this option with my new dual core 3800+ X2 at 2.75Ghz, and it was about 44% faster than my old opteron 144 at 2.68Ghz. other dual core optimized tools have also showed dramatic speed increases... i'm loving my new dual core =].

[n3r0]

i'm using intels e6600 and i have set dvdrebuilder to use multiple threads with CCE
but what i'm getting is 50-65% CPU load occasionally 70% max
is there any way i could get 100% CPU load

and multiple threads disabled the CPU load is mostly 70% occasionally 80% :eek: where's the catch:confused:

[nwg]

I still haven't bought a dual core since the only supplier I want to use it out of stock.

As I have a single core 3800 (San Deigo). Does it make any sense to get a Dual Core 3800 as they are easier to get hold off? I am thinking perhaps no as it runs at 2ghz per core rather than the 3800 2.2ghz. Or, can I expect a near 50% increase in speed with a X2 4200 with two 2.2ghz cores? I will not get into overclocking.

I want a noticeable speed increase using Procoder 2/CCE. The average Procoder encode take around 6 hours at High Quality (used to be 12 with a Athlon XP 2000). It would be nice to get close to 3 hours.

[blutach]

@nwg - I get almost real time encoding on PAL with a 3800 X2 no o/c.

Regards

[DDogg]

Abit AW8D mobo w/ Intel D930 OC'ed to 4.5 GHz (Clock @ 300). 2 gigs 667 ram 4-4-4-12.

Multicore = on. Using CCE 2.7 showed 6 x realtime and this process below took exactly 30 minutes including the ISO creation. Log below is QEnc with default settings. QEnc showed 2 instances, each encoding at 60 fps. Total process time 32 minutes.

If anybody wants to do a comparison with their hardware, the source was from a DVD I had laying around called Firewall. DVDFab dycrypter was used to place source on HD. Settings for QEnc - HQ checked and single pass cq vbr


[17:51:21] One Click encoding activated...
-----------------
[17:51:21] Phase I, PREPARATION started.
- DVD-RB v1.10.3
- QuEnc 0.7.1.0 encoder selected.
- "One Pass VBR (w/analysis)" mode is enabled.
- "Movie Only" mode is enabled.
- VTS_01: 2,855,999 sectors.
-- Scanning and writing .D2V & .AVS files
-- Processed 155,610 frames.
-- Building .AVS and .ECL files
- Reduction Level for DVD-5: 90.6%
- Overall Bitrate : 6,439/5,151Kbs
- Space for Video : 4,080,914KB
- Analyzing VTS_01 for optimal CQ factor.
-- TargetSize (sectors):2,448,548
-- Sampling 3120 of 155610 frames.
-- Predicted size (sectors) at CQ=2: 1,236,067
-- Predicted size (sectors) at CQ=1: 2,609,279
- CQ Value selected: 2
- HIGH/LOW/TYPICAL Bitrates: 6,222/4,658/5,151 Kbs
[17:55:45] Phase I, PREPARATION completed in 4 minutes.
[17:55:45] Phase II ENCODING started
- Creating M2V for VTS_01 segment 0
- Creating M2V for VTS_01 segment 1
- Creating M2V for VTS_01 segment 2
- Creating M2V for VTS_01 segment 3
- Creating M2V for VTS_01 segment 4
- Creating M2V for VTS_01 segment 5
- Creating M2V for VTS_01 segment 6
- Creating M2V for VTS_01 segment 7
- Creating M2V for VTS_01 segment 8
- Creating M2V for VTS_01 segment 9
- Creating M2V for VTS_01 segment 10
- Creating M2V for VTS_01 segment 11
- Creating M2V for VTS_01 segment 12
- Creating M2V for VTS_01 segment 13
- Creating M2V for VTS_01 segment 14
- Creating M2V for VTS_01 segment 15
- Creating M2V for VTS_01 segment 16
- Creating M2V for VTS_01 segment 17
- Creating M2V for VTS_01 segment 18
- Creating M2V for VTS_01 segment 19
- Creating M2V for VTS_01 segment 20
- Creating M2V for VTS_01 segment 21
- Creating M2V for VTS_01 segment 22
- Creating M2V for VTS_01 segment 23
- Creating M2V for VTS_01 segment 24
- Creating M2V for VTS_01 segment 25
- Creating M2V for VTS_01 segment 26
- Creating M2V for VTS_01 segment 27
[18:16:57] Phase II ENCODING completed in 21 minutes.
[18:16:57] Phase III, REBUILD started.
- Processing VTS_01
- Reading/processing TMAP table...
- Rebuilding seg 0 VOBID 5 CELLID 1
- Rebuilding seg 1 VOBID 5 CELLID 2
- Rebuilding seg 2 VOBID 5 CELLID 3
- Rebuilding seg 3 VOBID 5 CELLID 4
- Rebuilding seg 4 VOBID 5 CELLID 5
- Rebuilding seg 5 VOBID 5 CELLID 6
- Rebuilding seg 6 VOBID 5 CELLID 7
- Rebuilding seg 7 VOBID 5 CELLID 8
- Rebuilding seg 8 VOBID 5 CELLID 9
- Rebuilding seg 9 VOBID 5 CELLID 10
- Rebuilding seg 10 VOBID 5 CELLID 11
- Rebuilding seg 11 VOBID 5 CELLID 12
- Rebuilding seg 12 VOBID 5 CELLID 13
- Rebuilding seg 13 VOBID 5 CELLID 14
- Updating NAVPACKS for VOBID_05
- Rebuilding seg 14 VOBID 6 CELLID 1
- Rebuilding seg 15 VOBID 6 CELLID 2
- Rebuilding seg 16 VOBID 6 CELLID 3
- Rebuilding seg 17 VOBID 6 CELLID 4
- Rebuilding seg 18 VOBID 6 CELLID 5
- Rebuilding seg 19 VOBID 6 CELLID 6
- Rebuilding seg 20 VOBID 6 CELLID 7
- Rebuilding seg 21 VOBID 6 CELLID 8
- Rebuilding seg 22 VOBID 6 CELLID 9
- Rebuilding seg 23 VOBID 6 CELLID 10
- Rebuilding seg 24 VOBID 6 CELLID 11
- Rebuilding seg 25 VOBID 6 CELLID 12
- Rebuilding seg 26 VOBID 6 CELLID 13
- Rebuilding seg 27 VOBID 6 CELLID 14
- Updating NAVPACKS for VOBID_06
- Updated VTS_C_ADT.
- Updated VTS_VOBU_ADMAP.
- Updated IFO: VTS_01_0.IFO
- Correcting VTS Sectors...
- Building ISO Image...
- ISO Image successfully created.
[18:23:11] Phase III, REBUILD completed in 7 minutes.

Done.
[18:23:11] PREPARE/ENCODE/REBUILD completed in 32 min.

[jdobbs]

Wow. That sure makes my Athlon XP 3200 look bad...

I think I want a real computer when I grow up.

[blutach]

It was with CCE, which is especially quick, but still, I'll take DDogg's hand-me-downs :D I've seen DVD Shrink perform slower.

Regards

[jdobbs]

I get about 2.5x speed with CCE 2.70... about 2.7-3.0x with CCE 2.50 -- and up until now I thought I was smokin' :eek:

Unfortunately I have to limit myself to a middle-of-the-road computer. It's a bad idea to do your testing on a machine that's too fast. It'll make you lazy. I have to make sure I feel the pain whenever I do something that makes DVD-RB go slower.

But I sure would like to be a space cowboy like you guys...

[DDogg]

A few more data points from the above posted source and detail. These are just the encoding times using multicore and different encoders

CCE 2.7 - Phase II ENCODING completed in 17 minutes
QEnc - Phase II ENCODING completed in 21 minutes
HCenc - Phase II ENCODING completed in 23 minutes

@JD, I'm not up on this stuff anymore - While sure, MultiCore launches 2 instances of HC and QEnc, does it do anything for CCE? Or, has CCE been changed to better utilize duals? It hauls ashes. I know my machine is decent and all, but hell, I am seeing 7.5 x on low complexity source like titles and 6.25 to 6.50 on normal source. Gee, who would have thunk it!

[jdobbs]

I think CCE is multithreaded... I've tried getting two instances to run just to see what happens, but have been unsuccessful.

[casperse]

How does the DVDrebuilder utillize a Dual core like a AMD X2 4800+?

From your respons above I guess it dosent run 2 CCE but instead uses load sharing on 2 CPU running 1 CCE encoding process correct?

Sorry I only have one CPU so I cant test it....
(Getting one during this week...:D )

[jdobbs]

If the software supports multithreading then running two instances probably wouldn't help much (if any). If it only runs a single thread, though, two instances can improve throughput. It's also possible that for processors that are so fast the encoder doesn't push 100% utilization, and the encoder is single threaded, you might get an improvement as well.

[n3r0]

ok I tried to encode ACDC - stiff upper lip DVD with CCE one pass VBR
[14:59:18] One Click encoding activated...
-----------------
[14:59:19] Phase I, PREPARATION started.
- DVD-RB v1.10.3
- CCE 2.70.2.6 encoder selected.
- "One Pass VBR (w/analysis)" mode is enabled.
- Audio remapping mode is enabled.
- VTSM_01: 94 144 sectors.
-- Scanning and writing .D2V & .AVS files
-- Processed 6 080 frames.
-- Building .AVS and .ECL files
- VTS_01: 4 389 sectors.
-- Scanning and writing .D2V & .AVS files
-- Processed 270 frames.
-- Building .AVS and .ECL files
- VTS_09: 3 409 727 sectors.
-- Scanning and writing .D2V & .AVS files
-- Processed 229 443 frames.
-- Building .AVS and .ECL files
- VTS_10: 218 375 sectors.
-- Scanning and writing .D2V & .AVS files
-- Processed 19 070 frames.
-- Building .AVS and .ECL files
- Reduction Level for DVD-5: 65,1%
- Overall Bitrate : 3 759/3 007Kbs
- Space for Video : 3 902 380KB
- Analyzing VTS_01 for optimal Q factor.
-- TargetSize (sectors):2 890
-- Sampling 12 of 270 frames.
- Q Value selected: 10
- Analyzing VTS_09 for optimal Q factor.
-- TargetSize (sectors):1 784 433
-- Sampling 4590 of 229443 frames.
-- Predicted size (sectors) at Q=33: 3 257 007
-- Predicted size (sectors) at Q=71: 2 350 534
-- Predicted size (sectors) at Q=107: 1 829 466
-- Predicted size (sectors) at Q=112: 1 775 316
- Q Value selected: 112
- Analyzing VTS_10 for optimal Q factor.
-- TargetSize (sectors):132 786
-- Sampling 390 of 19070 frames.
-- Predicted size (sectors) at Q=37: 211 115
-- Predicted size (sectors) at Q=67: 149 793
-- Predicted size (sectors) at Q=80: 134 441
-- Predicted size (sectors) at Q=81: 133 452
- Q Value selected: 82
- HIGH/LOW/TYPICAL Bitrates: 4 217/2 282/3 007 Kbs
[15:06:44] Phase I, PREPARATION completed in 7 minutes.
[15:06:44] Phase II ENCODING started
- Creating M2V for VTSM_01 segment 0
- Creating M2V for VTSM_01 segment 1
- Creating M2V for VTSM_01 segment 2
- Creating M2V for VTSM_01 segment 3
- Creating M2V for VTSM_01 segment 4
- Creating M2V for VTSM_01 segment 5
- Creating M2V for VTSM_01 segment 6
- Creating M2V for VTS_01 segment 0
- Creating M2V for VTS_09 segment 0
- Creating M2V for VTS_09 segment 1
- Creating M2V for VTS_09 segment 2
- Creating M2V for VTS_09 segment 3
- Creating M2V for VTS_09 segment 4
- Creating M2V for VTS_09 segment 5
- Creating M2V for VTS_09 segment 6
- Creating M2V for VTS_09 segment 7
- Creating M2V for VTS_09 segment 8
- Creating M2V for VTS_09 segment 9
- Creating M2V for VTS_09 segment 10
- Creating M2V for VTS_09 segment 11
- Creating M2V for VTS_09 segment 12
- Creating M2V for VTS_09 segment 13
- Creating M2V for VTS_09 segment 14
- Creating M2V for VTS_09 segment 15
- Creating M2V for VTS_09 segment 16
- Creating M2V for VTS_09 segment 17
- Creating M2V for VTS_09 segment 18
- Creating M2V for VTS_09 segment 19
- Creating M2V for VTS_09 segment 20
- Creating M2V for VTS_09 segment 21
- Creating M2V for VTS_09 segment 22
- Creating M2V for VTS_09 segment 23
- Creating M2V for VTS_09 segment 24
- Creating M2V for VTS_10 segment 0
[15:36:26] Phase II ENCODING completed in 30 minutes.
[15:36:26] Phase III, REBUILD started.
- Copying IFO, BUP, and unaltered files...
- Processing VTS_01
- Reading/processing TMAP table...
- Rebuilding seg 0 VOBID 1 CELLID 1
- Updating NAVPACKS for VOBID_01
- Updated VTS_C_ADT.
- Updated VTS_VOBU_ADMAP.
- Updated IFO: VTS_01_0.IFO
- Updating TMAP table...
- Processing VTS_09
- Reading/processing TMAP table...
- Rebuilding seg 0 VOBID 1 CELLID 1
- Rebuilding seg 1 VOBID 1 CELLID 2
- Rebuilding seg 2 VOBID 1 CELLID 3
- Rebuilding seg 3 VOBID 1 CELLID 4
- Rebuilding seg 4 VOBID 1 CELLID 5
- Rebuilding seg 5 VOBID 1 CELLID 6
- Rebuilding seg 6 VOBID 1 CELLID 7
- Rebuilding seg 7 VOBID 1 CELLID 8
- Rebuilding seg 8 VOBID 1 CELLID 9
- Rebuilding seg 9 VOBID 1 CELLID 10
- Rebuilding seg 10 VOBID 1 CELLID 11
- Rebuilding seg 11 VOBID 1 CELLID 12
- Rebuilding seg 12 VOBID 1 CELLID 13
- Rebuilding seg 13 VOBID 1 CELLID 14
- Rebuilding seg 14 VOBID 1 CELLID 15
- Rebuilding seg 15 VOBID 1 CELLID 16
- Rebuilding seg 16 VOBID 1 CELLID 17
- Rebuilding seg 17 VOBID 1 CELLID 18
- Rebuilding seg 18 VOBID 1 CELLID 19
- Rebuilding seg 19 VOBID 1 CELLID 20
- Rebuilding seg 20 VOBID 1 CELLID 21
- Rebuilding seg 21 VOBID 1 CELLID 22
- Rebuilding seg 22 VOBID 1 CELLID 23
- Rebuilding seg 23 VOBID 1 CELLID 23
- Rebuilding seg 24 VOBID 1 CELLID 23
- Updating NAVPACKS for VOBID_01
- Audio_01 remapped to Audio_03
- Audio_03 remapped to Audio_01
- Updated VTS_C_ADT.
- Updated VTS_VOBU_ADMAP.
- Updated IFO: VTS_09_0.IFO
- Updating TMAP table...
- Processing VTS_10
- Reading/processing TMAP table...
- Rebuilding seg 0 VOBID 1 CELLID 1
- Updating NAVPACKS for VOBID_01
- Updated VTS_C_ADT.
- Updated VTS_VOBU_ADMAP.
- Updated IFO: VTS_10_0.IFO
- Updating TMAP table...
- Processing VTSM_01
- Rebuilding seg 0 VOBID 1 CELLID 1
- Updating NAVPACKS for VOBID_01
- Rebuilding seg 1 VOBID 2 CELLID 1
- Updating NAVPACKS for VOBID_02
- Rebuilding seg 2 VOBID 3 CELLID 1
- Updating NAVPACKS for VOBID_03
- Rebuilding seg 3 VOBID 4 CELLID 1
- Updating NAVPACKS for VOBID_04
- Rebuilding seg 4 VOBID 5 CELLID 1
- Updating NAVPACKS for VOBID_05
- Rebuilding seg 5 VOBID 6 CELLID 1
- Possible dropped frame(s) [ 91 ] noted in this segment, continuing...
- Updating NAVPACKS for VOBID_06
- Rebuilding seg 6 VOBID 7 CELLID 1
- Updating NAVPACKS for VOBID_07
- Updated VTSM_C_ADT.
- Updated VTSM_VOBU_ADMAP.
- Updated IFO: VTS_01_0.IFO
- Correcting VTS Sectors...
[15:53:39] Phase III, REBUILD completed in 17 minutes.

Done.
[15:53:39] PREPARE/ENCODE/REBUILD completed in 54 min.
mostly the speed was at 6x but what concerned me, was that CPU load was 65-80% multiple encoder processes enabled :confused:

it definitely could be better, if the CPU load was 95-100%
i closed most of the unneeded processes at the time and I'm running intel e6600 @ 2,7GHz with via 880pro chipset and pc3200 DDR memory

i hope the new version of CCE would be more efficiently optimized for dual core processors, cause v2,702 is not, at least not enough :p

[Nem]

Here are my results on Firewall with a Core 2 Duo E6400 at FSB 315x8 = 2.52GHz , ddr2-945 4.0-4-4-12 (all I can get on this crap AB9 Pro board). This is QuEnc, movie only, High Quality, Single Pass CQ VBR.


-----------------
[22:58:56] Phase I, PREPARATION started.
- DVD-RB v1.10.4
- QuEnc 0.7.1.0 encoder selected.
- "One Pass VBR (w/analysis)" mode is enabled.
- "Movie Only" mode is enabled.
- VTS_01: 2,855,999 sectors.
-- Scanning and writing .D2V & .AVS files
-- Processed 155,610 frames.
-- Building .AVS and .ECL files
- Reduction Level for DVD-5: 91.0%
- Overall Bitrate : 6,466/5,173Kbs
- Space for Video : 4,098,236KB
- Analyzing VTS_01 for optimal CQ factor.
-- TargetSize (sectors):2,458,942
-- Sampling 3120 of 155610 frames.
-- Predicted size (sectors) at CQ=2: 1,201,726
-- Predicted size (sectors) at CQ=1: 2,601,279
- CQ Value selected: 2
- HIGH/LOW/TYPICAL Bitrates: 6,249/4,678/5,173 Kbs
[23:06:13] Phase I, PREPARATION completed in 8 minutes.
[23:06:24] Phase II ENCODING started
- Creating M2V for VTS_01 segment 0
- Creating M2V for VTS_01 segment 1
- Creating M2V for VTS_01 segment 2
- Creating M2V for VTS_01 segment 3
- Creating M2V for VTS_01 segment 4
- Creating M2V for VTS_01 segment 5
- Creating M2V for VTS_01 segment 6
- Creating M2V for VTS_01 segment 7
- Creating M2V for VTS_01 segment 8
- Creating M2V for VTS_01 segment 9
- Creating M2V for VTS_01 segment 10
- Creating M2V for VTS_01 segment 11
- Creating M2V for VTS_01 segment 12
- Creating M2V for VTS_01 segment 13
- Creating M2V for VTS_01 segment 14
- Creating M2V for VTS_01 segment 15
- Creating M2V for VTS_01 segment 16
- Creating M2V for VTS_01 segment 17
- Creating M2V for VTS_01 segment 18
- Creating M2V for VTS_01 segment 19
- Creating M2V for VTS_01 segment 20
- Creating M2V for VTS_01 segment 21
- Creating M2V for VTS_01 segment 22
- Creating M2V for VTS_01 segment 23
- Creating M2V for VTS_01 segment 24
- Creating M2V for VTS_01 segment 25
- Creating M2V for VTS_01 segment 26
- Creating M2V for VTS_01 segment 27
[23:24:51] Phase II ENCODING completed in 18 minutes.


Also did CCE encoding time 15 minutes
And HC encoder took 25 minutes.

I might be faster encoding (at least in CCE and QuEnc anyway..) but your Phase I times blow my mind. I'm using a 4 drive raid 10 for source, 2 drive raid 0 for working folder, and destination folder on a single 250GB drive, and I average 7-10 minutes for Phase 1 on most stuff, and thats with CCE and no optimal CQ scan. Phase 3 is really slow for me too... takes average of 13 min to rebuild and pack to ISO.

I just completed 2 DVD's at my normal settings, CCE SP 2.7 3passes 43 minutes for FFVII Advent Children (130662 frames), 50 minutes for 50 cent Get Rich or Die Tryin (168066 frames). CCE gives almost steady speed of 7 most of the time (on first pass anyway, speed reading is borked on pass 2 and 3, but they take almost exactly the same ammount of time every time).

[n3r0]

Here are my results on Firewall with a Core 2 Duo E6400 at FSB 315x8 = 2.52GHz , ddr2-945 4.0-4-4-12 (all I can get on this crap AB9 Pro board). This is QuEnc, movie only, High Quality, Single Pass CQ VBR.

Also did CCE encoding time 15 minutes
And HC encoder took 25 minutes.

I might be faster encoding (at least in CCE and QuEnc anyway..) but your Phase I times blow my mind. I'm using a 4 drive raid 10 for source, 2 drive raid 0 for working folder, and destination folder on a single 250GB drive, and I average 7-10 minutes for Phase 1 on most stuff, and thats with CCE and no optimal CQ scan. Phase 3 is really slow for me too... takes average of 13 min to rebuild and pack to ISO.

I just completed 2 DVD's at my normal settings, CCE SP 2.7 3passes 43 minutes for FFVII Advent Children (130662 frames), 50 minutes for 50 cent Get Rich or Die Tryin (168066 frames). CCE gives almost steady speed of 7 most of the time (on first pass anyway, speed reading is borked on pass 2 and 3, but they take almost exactly the same ammount of time every time).
well one of the reason why the encoding took so much time for me, was that the source is about 140 minutes long and - Reduction Level for DVD-5 was 65,1%
but about that phase I
i have no clue why it is so fast :rolleyes:
the vob's and work folder were on the same physical drive but on different partitions
hitachi t7k250 160GB drive

[casperse]

I just got my AMD X2 4800+ (The new prices made this a cheap upgrade..A Intel would mean new MB, RAM, CPU-Cooler etc) today and the first run look very impressive here is the Stats on normal jobs I do and no OC! :

- Versions :
-- Rebuilder: v1.10.3
-- CCE SP Version: 2.70.2.4
-- ReJig Version: 0.5.0.5
-- QuEnc Version: 0.6.1.0
-- HC Version: 0.16.0.1
-- DGDECODE Version: 1.4.5
-- AVISYNTH Version: 2.5.6.0

Memoirs Of A Geisha Playing time 139 Min
CCE 4 passes + Menu encoding
-----------------
[22:16:14] Phase I, PREPARATION started.
- DVD-RB v1.10.3
- CCE 2.70.2.4 encoder selected.
- VTS_01: 2.490.477 sectors.
-- Scanning and writing .D2V & .AVS files
-- Processed 208.798 frames.
-- Building .AVS and .ECL files
- Reduction Level for DVD-5: 85,4%
- Overall Bitrate : 3.690Kbs
- Space for Video : 3.762.504KB
- HIGH/LOW/TYPICAL Bitrates: 5.737/371/3.690 Kbs
[22:18:01] Phase I, PREPARATION completed in 2 minutes.
[22:19:29] Phase II ENCODING started
- Creating M2V for VTS_01 segment 0
....................
- Creating M2V for VTS_01 segment 30
[00:34:35] Phase II ENCODING completed in 135 minutes.

*********************************************
V_FOR_VENDETTA PLAYING TIME 128 min
CCE 4 passes + Menu encoding

-----------------
[08:04:08] Phase I, PREPARATION started.
- DVD-RB v1.10.3
- CCE 2.70.2.4 encoder selected.
- VTSM_01: 26.899 sectors.
-- Scanning and writing .D2V & .AVS files
-- Processed 1.882 frames.
-- Building .AVS and .ECL files
- VTS_01: 2.945.066 sectors.
-- Scanning and writing .D2V & .AVS files
-- Processed 190.970 frames.
-- Building .AVS and .ECL files
- Reduction Level for DVD-5: 74,1%
- Overall Bitrate : 4.341Kbs
- Space for Video : 4.087.950KB
- HIGH/LOW/TYPICAL Bitrates: 8.552/3.149/4.341 Kbs
[08:06:32] Phase I, PREPARATION completed in 2 minutes.
[08:08:15] Phase II ENCODING started
- Creating M2V for VTSM_01 segment 0
..........
- Creating M2V for VTS_01 segment 34
[10:12:04] Phase II ENCODING completed in 124 minutes.

My normal encoding time before was normally around 250-300 min!!! :D
Now its about realtime (Playing time = Encoding) wich is great no more night jobs LOL

But remember that the temp on this is higher so you need a god airflow in the machine...with the bigger Zalman and full load I get a CPU temp around 50-52C which is normal.
Idle temp is around 44C but no errors its rock stable..So I can recommend this if you already have a Socket 939 and dont want to spend alot of money moving to Intel :sly:

[DDogg]

casperse, if you could encode Firewall in the 1 pass modes like Nem did, we would have an excellent comparison between the oc'ed Intel D's, Core2 Duos, and your AMD x2.

Nem, hmmm, my source is on striped array, working on striped array, final ISO and Video-TS back to the same array with the source.

I put some benchmarks here, if you have not already seen them on the Abit forum -
http://img104.imageshack.us/my.php?image=aw8dmf6.jpg

[nwg]

@nwg - I get almost real time encoding on PAL with a 3800 X2 no o/c.

Regards


Hi

I get an average of 0.40 - 0.70 x realtime with my Athlon64 3800 single core.

Does two encoding processes appear at once at nearly realtime each or, does the two cores just speed up the current and only encoding process to near realtime?

[casperse]

Sorry I am living in Denmark so Firewall is not on DVD yet (Region 2 Bullshit)

Yes it speeds up the one encoding process up on both CPU´s CCE. to around 70% load

Still you have to remember this is "Realtime" with passes= 4 so I think its pretty god. But I do think that the new Conroe can do even better but again this is a cheap upgrade compared to getting a new PC :D

[Sophoclesdrf]

I think CCE is multithreaded... I've tried getting two instances to run just to see what happens, but have been unsuccessful.

CCE is indeed multi threaded and having two cores while encoding with RB/CCE is an advantage during the encoding process. RB, DGDecode.dll, and Avisynth are not.

Note: If this is redundant then ignore me and move on.:)

[mezzanine]

69 minutes for both cores and 109 for single core with HC 0.18 using my awesome new core 2 duo E6400@3.2ghz. Movie is 101 minutes long!
:cool: :rolleyes: ;)

[casperse]

Does the Intel Conroe utillize 100 CPU when encoding?
Or does it use 70-80% like AMD?

Yes I have heard alot of god things about that new Intel CPU :D

[mezzanine]

Does the Intel Conroe utillize 100 CPU when encoding?
Or does it use 70-80% like AMD?

Yes I have heard alot of god things about that new Intel CPU :D

49 minutes for CCE SP 2.70

Well it's kinda strange it doesn't use 100% with all encoders under all apllications.
For example with DVD-RB and CCE it doesn't use more than 80% but when running CCE under DVD2DVD-R both cores are at 100%.

[blutach]

Hi

I get an average of 0.40 - 0.70 x realtime with my Athlon64 3800 single core.

Does two encoding processes appear at once at nearly realtime each or, does the two cores just speed up the current and only encoding process to near realtime?A process runs in each core. But it's not quite as simple as that as there is "crossover" when 1 processor has a slice to give the other.

Regards