Speed Poll for CCE users:

[apfraats]

Ok, I noticed some remarkable fast systems here.

I'm just running an Athlon 3000+ and want to compare speeds with other users.

We should consider:

1) No use of AVS filters.
2) No use of CCE filters.
3) No specific ballast.

So a staight compressing job at a 70%-80% level.

Let's say a 2 hour movie with 5.1 DD only.

I see remarkable speeds. If you can post youre CCE speed indication while it's doing second or third pass M2V , I would greatly appreciate it.

Also mention youre CCE version and system CPU/MEMORY/SPEED.

That because I see people doing jobs in such a short time, I really think I have to upgrade to another environment.


My speed:

1.8 using CCE 2.70.2.4 Athlon XP 3000+ 1GB memeory dual channel.

If others just could post something like this, I will get an overall impression.

A job can easily last for 4 hours using 3 pass CCE , UNDOT() and CCE low pass filter.

If it is possible to cut these time to 50% or less, it can be interesting to spend money on an upgrade to what I prefer: A dual Core Athlon 64 with 2* 1MB cache and 1GB or 2GB memory (well what does it costs anyway ?)

Oh anybody knows something about running DVD-RB with CCE on a full 64 bits enabled platform, such as WINDOWS XP 64 bits ???

Thanks a lot guys, give youre speeds and info please.

[Harrysmiith]

I use a Pentium 4 3.2 Gig 1 gig ram CCE Basic movie 129 mins
typical speed on 2nd pass was 2.6 to 2.7

took 96 mins to encode.

[18:22:55] Phase I, PREPARATION started.
- CCE Basic 2.70.1.4 encoder selected.
- "Movie Only" mode is enabled.
- VTS_01: 2,813,328 sectors.
-- Scanning and writing .D2V & .AVS files
-- Processed 192,952 frames.
-- Building .AVS and .ECL files
- Reduction Level for DVD-5: 77.9%
- Overall Bitrate : 4,315Kbs
- Space for Video : 4,064,942KB
- HIGH/LOW/TYPICAL Bitrates: 4,581/3,429/4,315 Kbs
[18:29:18] Phase I, PREPARATION completed in 7 minutes.
[18:32:07] 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
[20:08:42] Phase II ENCODING completed in 96 minutes.

[DD51]

I keep a copy of all my encodes.
PC Specs:
P4 3.2E overclocked @ 3.9GHZ
1GIG Memory
CCE SP 2.70.2.4
Avg. Speed between 4.2 to 4.6

Here's my latest encode:
The 40 Year Old Virgin
[17:44:32] One Click encoding activated...
-----------------
[17:44:32] Phase I, PREPARATION started.
- CCE SP 2.70.2.4 encoder selected.
- "Movie Only" mode is enabled.
- VTS_01: 2,742,399 sectors.
-- Scanning and writing .D2V & .AVS files
-- Processed 190,547 frames.
-- Building .AVS and .ECL files
- Reduction Level for DVD-5: 79.7%
- Overall Bitrate : 5,118/4,095Kbs
- Space for Video : 3,972,254KB
- HIGH/LOW/TYPICAL Bitrates: 4,314/393/4,095 Kbs
[17:46:39] Phase I, PREPARATION completed in 2 minutes.
[17:46:39] 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
- Extracting STILLS for VTS_01 segment 21
[18:46:20] Phase II ENCODING completed in 60 minutes.

[jptheripper]

on my homepc p3 1ghz 1.5gb ram - average speed 0.6 (~30hours per job)
on my workpc p4 3.73EE 2gb ram - average speed 3.8 (~2 hours per job)

[archaeo]

Dual PIII xeon's @ 1 ghz, 1 gb ram, CCE 2.70.2 speed is about 1.4 on average

[jdobbs]

AMD XP 3200+, 512MB Dual Channel:

CCE Basic 2.70.01.07, 2.25x - 2.50x
CCE SP v2.50, 2.50x - 2.90x
CCE SP Trial v2.70.01.05, 2.20x - 2.30x

[l8nights]

p4 @3ghz 512, cce 2.70 roughly average 2.8x-3x w/ anything under 7gbs 45-50min encode 70-75 min total building iso's

cheers!

[markrb]

AMD X2 3800+ slightly overlocked
2gb ram

CCE 2.7.02.04 4-4.2
CCE 2.51 4-4.8

Each encode is a little different.

Mark

[dozkid]

AMD XP 2200+, 256MB ram
CCE SP Trial v2.70.02.04
roughly average 1.6x - 1.7x

[hallway]

I use a Pentium 4 3.2 Gig 1 gig ram CCE Basic movie 129 mins

- CCE Basic 2.70.1.4 encoder selected.
- "Movie Only" mode is enabled.
- Reduction Level for DVD-5: 77.9%
[20:08:42] Phase II ENCODING completed in 96 minutes.

P4 3.2E overclocked @ 3.9GHZ
1GIG Memory
CCE SP 2.70.2.4

- CCE SP 2.70.2.4 encoder selected.
- "Movie Only" mode is enabled.
- Reduction Level for DVD-5: 79.7%
[18:46:20] Phase II ENCODING completed in 60 minutes. Those look very similar hardware wise and compression amount wise. Is overclocking making such a difference or is it the encoder, basic vs SP ??

[apfraats]

Thanks all, for info, keep it comming !

OK, let's go on, especially if someone out there uses Athlon XP 32 bits , as I am and the new X2's .


I see considerable speed increases up to more than 2 times......

Also Pentiums seems to be faster then Athlons 64.
The same pentium seems significantly faster with 1GB instaed of 512MB.

But they are too few results to make a statement.

So I'm not so happy with my 1.8 speed in M2V passes.:devil: :devil:


I was already considering building a system with DUAL CORE ATHLON XP 4400 DUAL 1MB CACHE, but that CPU is pretty high priced here (about 619 Euro, even more dollars, that I consider pretty much).

I saw 1 X2 dual core ATHLONS so far, so I'll wait and see what's comming up.

That annoying AMD is always messing around with clock speeds that aren't for real and now even more with the X2 series. I suspect them to translate the extra CPU power of the second core to semi clockspeed indication.

So a X2 4400 is in fact two 3500 Cores with each 1 MB cache, I see now that increased cache also is translated into their semi clockspeed indication......wonderfull wonderfull AMD, just another few months or so and they even translate internal memory to clock speeds.......

AMD get real !!! (what a mess they make of it).

So I have to compare at price level.

Please report CPU cache size also, as AMD uses 1MB cache as a manner to increase their clockspeed indication.

It looks that CCE loves pentiums SSE2, that the ATHLONS don't have.

But for some reason I do not wanna have a INTEL chip in my PC (yep, gonna see a shrink on this:o ).

Well keep the reports comming, and mention specialities like the 1 MB cache and X2 speed reports with the AMD LABELED clockspeed identification (not real clockspeed that's lower).

Keep them comming, let's see if I continue to love AMD....

Thanks

[hallway]

So I'm not so happy with my 1.8 speed in M2V passes. I seem to recall getting those kinds of speeds from CCE and I've only got an Athlon XP 2500+. That annoying AMD is always messing around with clock speeds that aren't for real and now even more with the X2 series. I suspect them to translate the extra CPU power of the second core to semi clockspeed indication. AMD has been using a "PR" rating for what, 10 years now ?? This is nothing new. The chip's marketing name, i.e. 4400 or 3500, implies that it's (roughly) equal to a Pentium running at 4.4ghz or 3.5ghz, respectively. Please report CPU cache size also, as AMD uses 1MB cache as a manner to increase their clockspeed indication.

It looks that CCE loves pentiums SSE2, that the ATHLONS don't have. The size of the CPU's cache likely plays a huge role. I remember when Intel came out with the original Xeon's, back in the 500mhz range, but they had 1-2mb of cache vs maybe 256-512k that the regular Pentiums had. The Xeons smoked the others by a wide margin.

AMD's current CPUs support SSE2. Not sure how far back they go though. I know my Athlon does not, only SSE(1).

[apfraats]

Yep I know what AMD tries us to tell:

"Please Please Please buy our CPU's as they are faster and cheaper then Intel"


Well based on what ?

Average performance ???
Floating point performance ???
Doing a sort in Word ???
Doing die hard video processing (like CCE) ???
Playing the latest 8D games with 12.5 surround ?

I suppose GENERAL, so overall performance.

BUT

We are dealing here with CCE (nice to hear SSE2 now is also supported !, the Ahtlon XP doens't do that, and I thought it was a law kind of issue, but it seems not the cae anymore, happy to know, thanx HALLWAY !)

I have red a lot and VIDEO processing seems to be the power of P4 Cpu's.

Probably because the instructions used are quicker executed having a higher basic clock, and AMD's more efficient microcode cannot compete this.

So now the question, and I LOVE TO SEE RESULTS is:

Comparing the ATHLON 4400 X2 2*1MB cache, to PENTIUM D830 3.0 GHZ with 2 cores also and also 2*1MB case.

Why ? Because I have the strange feeling the PENTIUM D830 3.0 GHZ is a lot faster then my lovely (wanting to buy) ATHLON 64 X2 4400 2*1MB cache.

If it were for INDICATED clockspeed the ATHLON should win.

But because this is video-stuff we are talking about, I'm afraid, much afraid that the PENTIUM D830 wins without any discussion.

The pentium D830 costs: 389 Euro
which is much cheaper then the
ATHLON 64 X2 4400 with it's 619 Euro price tag.

You can even buy the PENTIUM D840 for about the same price as the Athlon.

But: Basic clock of Athlon is just 2200 Mhz.
Basic clock of the PENTIUM is 3000 or even 3200 Mhz.

If we continue to compare PENTIUM IV to ATHLON XP , I'm afraid the PENTIUM D830 wins with great ease using CCE as a benchmark.

So the choice would be simple:

A faster PENTIUM D830 for 239,- less !!! (AAARggggghhhhhhhhh..)

And I have to visit my shrink because I need Intel all over (chipset e.d.) :eek: )

Any of you, although having the lack of money to buy a second original, :D :D , can effort already a fully loaded X2 4400+ or P830 system ???

[jptheripper]

well im building a system for a p4 830, but its a few months away budget wise

currently have:
asus 945p-sli deluxe motherboard
2gb 533mhx ram
x600pro all-in-wonder pcie card
1tb storage

need:
antec trueblue 480watt powersupply
antec p160w case
p4 830 dual 3.0 cpu

[Jeffster]

No-one has mentioned the source format (PAL/NTSC) when quoting their speed figures, but it can make big difference when you are comparing systems with similar specs. Maybe it's because my system is not the 'latest and greatest' I notice it more than others...

I have a celeron 1 ghz w/ 512 SDRAM - average speed 0.6 (PAL)

These figures are similar to the results jptheripper quoted for his home machine.

BUT on the few ocassions I've run NTSC backups I get an average speed of 0.75..... that's a 20% difference!

I guess most of that can be accounted for by the higher resolution of PAL and therefore the extra information to process.

[l8nights]

well im building a system for a p4 830, but its a few months away budget wise
why not spend the extra money and go for the 930
here are some reviews => http://www.xbitlabs.com/articles/cpu/display/presler.html

I found the second paragraph of page 9 of great interest!

where the 800 series (smithfield?) had some short comings the 900 has addressed them nicely or so I read!!!

hey what's a couple weeks?

[archaeo]

However I had read (Tom's Hardware Reviews) that the performance increase from the 800 to 900 series was only about 7%.

[geezer]

Have you seen this review?
http://www.tomshardware.com/cpu/20051121/the_mother_of_all_cpu_charts-31.html
(esp the video section)

BTW:
AMD X2 4200+
2 GB RAM
2 74GB WD Raptors
No o/c
CCE 2.70.02.04 Typ ~4.8 some sources as high as 5.1 (NaN)
Compression 80% no filters.

[jdobbs]

Wow... I wanna be just like you when I grow up... 4.8x :eek:

[geezer]

And I'm able to cruise the 'net (as well as other things) at the same time. CPU usage is ~75% with CCE as I'm doing now... No need for lower priority.

:^)

[Harrysmiith]

Those look very similar hardware wise and compression amount wise. Is overclocking making such a difference or is it the encoder, basic vs SP ??


I don't know but am happy to exchange machines in the interests of scientific enquiry :rolleyes:

[l8nights]

However I had read (Tom's Hardware Reviews) that the performance increase from the 800 to 900 series was only about 7%
I also read that review and it seemed if he had gotten an early release that didn't have the ht or other various capabilities! if yo umean this thread => http://www.tomshardware.com/2006/01/05/the_65_nm_pentium_d_900s_coming_out_party/page20.html

but it's something I thought would be worth some consideration!
hell tax time is right around the corner here may be able to see what I can squeeze out of one if the wife don't squeeze me out!

[random asshat]

http://images5.theimagehosting.com/encode.jpg

\ loves his AMD FX

[hallway]

Out of curiosity, I compared my work machine to my home machine and was really surprised at the results. This is the same movie, using the HC encoders, set on "Best" quality.

AMD Athlon XP 2500+, 1.5gb RAM, WinXP SP2
-----------------
[19:16:26] Phase I, PREPARATION started.
- HC encoder selected
- "Movie Only" mode is enabled.
- VTS_02: 2,565,682 sectors.
-- Scanning and writing .D2V & .AVS files
-- Processed 149,892 frames.
-- Building .AVS and .ECL files
- Reduction Level for DVD-5: 95.2%
- Overall Bitrate : 6,554/5,244Kbs
- Space for Video : 4,001,604KB
- HIGH/LOW/TYPICAL Bitrates: 6,149/4,590/5,244 Kbs
[19:24:59] Phase I, PREPARATION completed in 8 minutes.

... snip ...

- Creating M2V for VTS_02 segment 27
- Creating M2V for VTS_02 segment 28
- Creating M2V for VTS_02 segment 29
[23:35:49] Phase II ENCODING completed in 230 minutes.

================================================

Intel Pentium IV 2.4ghz (oc'd to 2.52ghz), 800mhz FSB, 2.00gb RAM, WinXP SP2
-----------------
[15:34:58] Phase I, PREPARATION started.
- HC encoder selected
- "Movie Only" mode is enabled.
- VTS_02: 2,565,682 sectors.
-- Scanning and writing .D2V & .AVS files
-- Processed 149,892 frames.
-- Building .AVS and .ECL files
- Reduction Level for DVD-5: 95.2%
- Overall Bitrate : 6,554/5,244Kbs
- Space for Video : 4,001,604KB
- HIGH/LOW/TYPICAL Bitrates: 6,149/4,590/5,244 Kbs
[15:42:49] Phase I, PREPARATION completed in 8 minutes.

... snip ...

- Creating M2V for VTS_02 segment 27
- Creating M2V for VTS_02 segment 28
- Creating M2V for VTS_02 segment 29
[21:09:11] Phase II ENCODING completed in 272 minutes.

[apfraats]

So the latest pentium or AMD ?

[Sophoclesdrf]

AMD dual core Opteron 175@2.70 Ghz
Memory timings 2-3-2-7 1T
2 gigs Corsair Twinx2048-3500LL Pro

I've had a lot of experience encoding with both AMD systems and Intel systems and I see some claims that are unrealistic. I think that every claim should include a screen shot of RB with the results displayed.

[jdobbs]

So what speeds are you seeing, Soph?

[hank315]

Intel P4 3.2 @ 3.36 - 1 GB RAM - 800 MHz FSB - WIN XP SP1

Movie: Matrix Revolutions, run time 124 min.

CCE basic
---------

[20:04:09] Phase I, PREPARATION started.
- CCE Basic 2.70.1.4 encoder selected.
- "Movie Only" mode is enabled.
- VTS_01: 2,955,451 sectors.
-- Scanning and writing .D2V & .AVS files
-- Processed 194,544 frames.
-- Building .AVS and .ECL files
- Reduction Level for DVD-5: 82.8%
- Overall Bitrate : 4,245Kbs
- Space for Video : 4,032,244KB
- HIGH/LOW/TYPICAL Bitrates: 5,184/4,021/4,245 Kbs
[20:08:38] Phase I, PREPARATION completed in 4 minutes.
[20:08:38] Phase II ENCODING started
- Creating M2V for VTS_01 segment 0
- Creating M2V for VTS_01 segment 1
- ......
- Creating M2V for VTS_01 segment 32
- Creating M2V for VTS_01 segment 33
[21:32:40] Phase II ENCODING completed in 84 minutes.


HC encoder (normal setting)
---------------------------

[16:49:39] Phase I, PREPARATION started.
- HC encoder selected
- "Movie Only" mode is enabled.
- VTS_01: 2,955,451 sectors.
-- Scanning and writing .D2V & .AVS files
-- Processed 194,544 frames.
-- Building .AVS and .ECL files
- Reduction Level for DVD-5: 82.8%
- Overall Bitrate : 4,245Kbs
- Space for Video : 4,032,244KB
- HIGH/LOW/TYPICAL Bitrates: 5,184/4,021/4,245 Kbs
[16:51:54] Phase I, PREPARATION completed in 2 minutes.
[16:51:54] Phase II ENCODING started
- Creating M2V for VTS_01 segment 0
- Creating M2V for VTS_01 segment 1
- ......
- Creating M2V for VTS_01 segment 32
- Creating M2V for VTS_01 segment 33
[19:16:16] Phase II ENCODING completed in 145 minutes.

[Sophoclesdrf]

jdobbs

I've done only about a half dozen movies and they of course vary. My encode speeds with CCE are about 5.70 to 6.20 the encode times have ranged between 48 minutes and 62 minutes. When I get things to where I want them I'll posts some screen shots.


I've been playing around with HC encoder lately because it frees up a core so that I can still surf the new without seeing any delay caused by encoding.

[jobe228]

Average speed started at 2.6 with the ILVU then went to 3.0-3.2x. I didn't watch too much with 10 pass. I know, I know, but a great movie!!. System is P4 3.0 ghz, 1GB RAM, 2-150GB Raptors Raid0, 120GB source drive. I want to upgrade to athlon 64 x2 when I have the time and money. Oh well.

[18:02:25] One Click encoding activated...
-----------------
[18:02:25] Phase I, PREPARATION started.
- CCE SP 2.70.2.4 encoder selected.
- "Half-D1/Half Space for Extras" mode is enabled.
- "Steal Space from Extras" mode is enabled.
- VTS_01: 2,405,457 sectors.
-- ANGLE and/or INTERLEAVING is present.
-- Scanning and writing .D2V & .AVS files
- Processed 248.9MB ILVU section...
- Processed 73.1MB ILVU section...
- Processed 491.4MB ILVU section...
-- Processed 147,620 frames.
-- Building .AVS and .ECL files
- VTS_02: 847,586 sectors.
-- Scanning and writing .D2V & .AVS files
-- Processed 56,782 frames.
-- Building .AVS and .ECL files
- Reduction Level for DVD-5: 61.3%
- Overall Bitrate : 4,351/3,481Kbs
- Space for Video : 3,622,728KB
- Movie improvement from extra reduction = 20.6%
- HIGH/LOW/TYPICAL Bitrates: 5,129/393/3,481 Kbs
[18:12:42] Phase I, PREPARATION completed in 10 minutes.
[18:12:42] Phase II ENCODING started
- Creating M2V for VTS_01 segment 0
[edit]
- Creating M2V for VTS_01 segment 24
- Creating M2V for VTS_02 segment 0
[edit]
- Creating M2V for VTS_02 segment 10
- Extracting STILLS for VTS_02 segment 11
[03:36:50] Phase II ENCODING completed in 564 minutes.
[03:36:50] Phase III, REBUILD started.
- Copying IFO, BUP, and unaltered files...
- Processing VTS_01
- Reading/processing TMAP table...
- Processed 248.9MB ILVU section...
- Rebuilding seg 0 VOBID 1 CELLID 1
- Rebuilding seg 1 VOBID 2 CELLID 1
- Updating NAVPACKS for VOBID_01
- Updating NAVPACKS for VOBID_02
- Rebuilding seg 2 VOBID 3 CELLID 1
- Rebuilding seg 3 VOBID 3 CELLID 2
- Rebuilding seg 4 VOBID 3 CELLID 3
- Rebuilding seg 5 VOBID 3 CELLID 4
- Processed 73.1MB ILVU section...
- Updating NAVPACKS for VOBID_03
- Rebuilding seg 6 VOBID 4 CELLID 1
- Rebuilding seg 7 VOBID 5 CELLID 1
- Updating NAVPACKS for VOBID_04
- Updating NAVPACKS for VOBID_05
- Rebuilding seg 8 VOBID 6 CELLID 1
- Rebuilding seg 9 VOBID 6 CELLID 2
- Rebuilding seg 10 VOBID 6 CELLID 3
- Rebuilding seg 11 VOBID 6 CELLID 4
- Rebuilding seg 12 VOBID 6 CELLID 5
- Rebuilding seg 13 VOBID 6 CELLID 6
- Rebuilding seg 14 VOBID 6 CELLID 7
- Rebuilding seg 15 VOBID 6 CELLID 8
- Rebuilding seg 16 VOBID 6 CELLID 9
- Rebuilding seg 17 VOBID 6 CELLID 10
- Rebuilding seg 18 VOBID 6 CELLID 11
- Rebuilding seg 19 VOBID 6 CELLID 12
- Processed 491.4MB ILVU section...
- Updating NAVPACKS for VOBID_06
- Rebuilding seg 20 VOBID 7 CELLID 1
- Rebuilding seg 21 VOBID 7 CELLID 2
- Rebuilding seg 22 VOBID 8 CELLID 1
- Rebuilding seg 23 VOBID 8 CELLID 2
- Updating NAVPACKS for VOBID_07
- Updating NAVPACKS for VOBID_08
- Rebuilding seg 24 VOBID 9 CELLID 1
- Updating NAVPACKS for VOBID_09
- Updated VTS_C_ADT.
- Updated VTS_VOBU_ADMAP.
- Updated IFO: VTS_01_0.IFO
- Updating TMAP table...
- Processing VTS_02
- Reading/processing TMAP table...
- 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
- Updating NAVPACKS for VOBID_06
- Rebuilding seg 6 VOBID 7 CELLID 1
- Updating NAVPACKS for VOBID_07
- Rebuilding seg 7 VOBID 8 CELLID 1
- Updating NAVPACKS for VOBID_08
- Rebuilding seg 8 VOBID 9 CELLID 1
- Updating NAVPACKS for VOBID_09
- Rebuilding seg 9 VOBID 10 CELLID 1
- Updating NAVPACKS for VOBID_10
- Rebuilding seg 10 VOBID 11 CELLID 1
- Updating NAVPACKS for VOBID_11
- Rebuilding seg 11 VOBID 12 CELLID 1
- Updating NAVPACKS for VOBID_12
- Updated VTS_C_ADT.
- Updated VTS_VOBU_ADMAP.
- Updated IFO: VTS_02_0.IFO
- Updating TMAP table...
- Correcting VTS Sectors...
[03:49:21] Phase III, REBUILD completed in 13 minutes.

[samuelal]

Hello.

I would also like to add to this poll.

Here is a snapshot from one of my backup projects:
http://img517.imageshack.us/img517/4936/ccedualcoretest0zk.th.jpg (http://img517.imageshack.us/my.php?image=ccedualcoretest0zk.jpg)

Watch the image closely - I captured it on the last encoding second on the last pass (using 6 passes) so the speed will get as close as possible to the real one.
* To view the image full screen you might have to click on it several times for it to fully resize at the 1152x864, 1:1 ratio it was taken.

Basically, the speed is around x5.20-x5.40

These are the application specs I was using while doing the backup:
CCE SP Trial v2.70.02.00
iDCT=5 (32bit SSE2/MMX)
NAN's DGDecode.dll v1.1.0+ SSE2 (http://forum.doom9.org/showthread.php?t=101353)
AviSynth v2.5.5
PAL Source @ 25fps
Movie Only:
LoadPlugin("C:\Program Files\DVD-RB\DGDecode.dll")
mpeg2source("C:\RB\DVD-TEST-WORK\D2VAVS\V01.D2V",idct=5)
trim(0,3206)
ConvertToYV12()

These are my computer specs while doing the backup:
Was using DVD-RB Pro v1.05.1
DFI LANParty UT nF4 SLI-DR Expert, Rev. A02 BIOS 11/02/2005
AMD Athlon64 X2 3800+ 2000MHz @ 2500MHz
Zalman CNPS9500 LED
OCZ EL PC-4000 2x1024MB Gold Ed. Kit @ 3-4-3-7-1T-7Trc-12Trfc-256Idle @ 500MHz 1:1 @ 2.71v
2 x LeadTek WinFast PX7800 GTX THD MyVIVO @ 490/1300
Creative Sound Blaster X-Fi Elite Pro
1 x WDC SE 200GB 8MB 7200 SATA I WD2000JD
2 x WDC SE 160GB 8MB 7200 SATA II WD1600JS @ RAID-0
Lite-On DVD/CD-RW SOHC5232K
NEC DVD-RW ND-4550A
Antec Performance One P160
Enermax Liberty 620W ELT620AWT


Hope this helps.

Samuel.

[apfraats]

I agree, it's a bit difficult to see the truth here.........

I always thought AMD was not in for 'video processing' and pentoum was.

If someone throws an AMD FX specification to me, I bankruppted already....

Just post dual core ATHLON XP and the new DUAL CORE PENTIUMS .

That's what the decision is about......

Not results using NASA's supercomputer..... :)

[Penecho]

I have AMD Athlon X2 4400+ (2200MHZ) @ 2500MHZ, 1 MB Cache/ each Core, 2 GB RAM DualChannel.

CCE 2.67 about 4x speed!!




Cu

[Rockas]

Basically, the speed is around x5.20-x5.40

These are the application specs I was using while doing the backup:
CCE SP Trial v2.70.02.00

Seems to me that you may be wrong there... not sure though... but I think that version of CCE only displays the correct speed whan making the VAF (first pass) file.

[samuelal]

Seems to me that you may be wrong there... not sure though... but I think that version of CCE only displays the correct speed whan making the VAF (first pass) file.

Hello!!! (Love your DVD-RB skin!)

I don't know man, That's why in the process of the test I waited until the last remaining second of the last (6th) pass - the speed sorta 'regulates' itself as time goes by up till the last pass.

Samuel.

[Rockas]

Hello!!! (Love your DVD-RB skin!)
Thanks 8)

As I said... I'm not sure... but i recall someone saying something about that... maybe that someone can confirm this :)

[Rippraff]

You're right Rockas. Speed display is fixed in version 2.70.2.4

Cu Rippraff

[jdobbs]

I think it's right when it gets to the end... it just seems to start each pass wrong.

[Rockas]

I think it's right when it gets to the end... it just seems to start each pass wrong.
Ummm... i don't know about that... I made a quick test and I get an avrg of 2.9 on the vaf and then I get 5.8 (yeah I wish) doing the m2v :confused:
Anyway... I check that out on a full backup.

[samuelal]

To: apfraats, Rockas, Ripraff & jdobbs: :)

I've decided to 'take on the challenge' and check it.

A few variables have changed since the last test I've done, so it might not be exactly apples-to-apples but I think they are quite small in the general way of things:
1. Fresh OS install, and for that matter:
A. AVISynth is now v2.5.6a instead of v2.5.5
B. Using DVD-RB Pro v1.06 with latest DGDecode v1.4.5 it supports, instead of DVD-RB Pro v1.05.1 with NAN's DGDecode 1.1.0+ SSE2 .
C. I've installed the CCE SP Trial v2.70.00.04.

Let's get down to buisness:
These are the application specs I was using while doing the backup:
CCE SP Trial v2.70.02.04
iDCT=3 (32bit SSE2/MMX) // Note: in DGDecode 1.4.5 I had to select 3 instead of 5 because of the differences in the iDCT's numbering.
DGDecode.dll v1.4.5
AviSynth v2.5.6a
PAL Source @ 25fps
Movie Only:
Code:

LoadPlugin("C:\Program Files\DVD-RB\DGDecode.dll")
mpeg2source("C:\RB\DVD-TEST-WORK\D2VAVS\V01.D2V",idct=5)
trim(0,3206)
ConvertToYV12()


Here is the new screenshot from the same backup project:
http://img300.imageshack.us/img300/9971/ccesptrialdualcoretest27000047.th.jpg (http://img300.imageshack.us/my.php?image=ccesptrialdualcoretest27000047.jpg)
Watch the image closely - I captured it on the last encoding second (again) on the last pass (using 6 passes) so the speed will get as close as possible to the real one.
* To view the image full screen you might have to click on it several times for it to fully resize at the 1152x864, 1:1 ratio it was taken.


These are my computer specs while doing the backup (have not changed from my previous test):
Was using DVD-RB Pro v1.06
DFI LANParty UT nF4 SLI-DR Expert, Rev. A02 BIOS 11/02/2005
AMD Athlon64 X2 3800+ 2000MHz @ 2500MHz
Zalman CNPS9500 LED
OCZ EL PC-4000 2x1024MB Gold Ed. Kit @ 3-4-3-7-1T-7Trc-12Trfc-256Idle @ 500MHz 1:1 @ 2.71v
2 x LeadTek WinFast PX7800 GTX THD MyVIVO @ 490/1300
Creative Sound Blaster X-Fi Elite Pro
1 x WDC SE 200GB 8MB 7200 SATA I WD2000JD
2 x WDC SE 160GB 8MB 7200 SATA II WD1600JS @ RAID-0
Lite-On DVD/CD-RW SOHC5232K
NEC DVD-RW ND-4550A
Antec Performance One P160
Enermax Liberty 620W ELT620AWT

To Rockas and Ripraff: I am pretty sure you guys weren't 100% right.
Look at the total time it took to encode:
1. Previous test was 2:25 mins and at a speed of about x5.23 at the end.
2. New test is 3:05 mins and at a speed of about x4.14 at the end.
Now this means that my findings regarding the speed were correct the first time and are also correct now (under the new circumstances).
I will do another test now using DVD-RB Pro 1.06 and NAN'S DGDecode v1.1.0+ SSE2 .


To jdobbs: You are also right - I haven't seen a version of CCE which doesn't slowly 'regulate' the total encoding speed which is being presented to the user up until now.

For instance, in this test I paid close attention to the first .VAF file creation process and saw that it started at x3.96, and when it reached the last second of the last (6th) pass of the encoding session - it reached x4.14 as shown in the new screenshot.

Thanks again guys for the heads up !

Samuel.

[samuelal]

For those who missed my last test update on this thread 40 mins ago: http://forum.doom9.org/showthread.php?p=771430#post771430

Ok guys, now here's the latest:

These are the application specs I was using while doing the backup:
CCE SP Trial v2.70.02.04
iDCT=5 (32bit SSE2/MMX)
NAN'S DGDecode.dll v1.1.0+ SSE2
AviSynth v2.5.6a
PAL Source @ 25fps
Movie Only:
Code:
LoadPlugin("C:\Program Files\DVD-RB\DGDecode.dll")
mpeg2source("C:\RB\DVD-TEST-WORK\D2VAVS\V01.D2V",idct=5)
trim(0,3206)
ConvertToYV12()

http://img10.imageshack.us/img10/7903/ccesptrialdualcoretest2700204n1.th.jpg (http://img10.imageshack.us/my.php?image=ccesptrialdualcoretest2700204n1.jpg)
Watch the image closely - I captured it on the last encoding second (again) on the last pass (using 6 passes) so the speed will get as close as possible to the real one.
* To view the image full screen you might have to click on it several times for it to fully resize at the 1152x864, 1:1 ratio it was taken.

Process time: 2:55 mins.
Speed at the last second of the last (6th) pass: x4.36
A change, of about 3.5-4% in favour of NAN's DGDecode v1.1.0 SSE2.
I will have to check his newest DGDecode v1.1.1 SSE2 ReRelease as well... interesting.

More interesting to me is that using the latest DGDecode (v1.4.5) while making sure it's using the same SSE2/MMX instruction path produces slower results.

This can hold to the CCE SP Trial v2.70.02.00 situation in such a way that:
1. CCE SP Trial v2.70.02.00's encoding engine is different and faster than 2.70.02.04's .
2. AVISynth v2.5.5 is faster than AVISynth v2.5.6a
3. DVD-RB Pro v1.05.1's encoding process is faster than DVD-RB Pro v1.06 - very unlikely since as far as I know the entire encoding process is done "outside" of the program.
4. CCE SP Trial v2.70
I dont' know about everything I've written here, just ruling every variable that changed on my part out.
Samuel.

P.S. sorry for all of this, seems to me this is all getting too close to hi-jacking the thread. My apologies to apfraats.

[random asshat]

I agree, it's a bit difficult to see the truth here.........

I always thought AMD was not in for 'video processing' and pentoum was.

If someone throws an AMD FX specification to me, I bankruppted already....

Just post dual core ATHLON XP and the new DUAL CORE PENTIUMS .

That's what the decision is about......

Not results using NASA's supercomputer..... :)


Sorry, that was me with the FX. I must have missed the part
about single core exclusion.

From my understanding the Athlon64's that clock the highest get
some more cache and become FX's

specs here (http://www.amd.com/us-en/Processors/ProductInformation/0,,30_118_9485_9488,00.html)

[Sophoclesdrf]

I stated in an earlier post that if you want to post your speeds then you should post them with a screen shot that removes the chance for altering stats. Here is a screen shot of a movie that I've recently encoded. I'll post more but I think that to keep things straight we should be doing the same movie with the same options.


http://img295.imageshack.us/img295/8287/rebuild15vx.jpg

[Buckwheat2005]

I have AMD Athlon X2 4400+ (2200MHZ) @ 2500MHZ, 1 MB Cache/ each Core, 2 GB RAM DualChannel.

CCE 2.67 about 4x speed!!

Cu

I'm getting the same speed for CCE SP v2.70.02.04.
speed 4.1 to 4.3.

AMD Athlon X2 4400+ @ 2.2 Ghz no 'oc.
2Gb Ram.

[Sophoclesdrf]

I'm getting some bizarre speeds with 2.70.2.1 which I just updated to. The first pass is normal anywhere from 5.4 to over 6 times but the second pass kicks into hyper drive. The first pass took over 2 minutes but the second pass took less than one minute.

It was in fact a bug in the version of CCE that I was using.

[SpazzHH]

I'm getting some bizarre speeds with 2.70.2.1 which I just updated to. The first pass is normal anywhere from 5.4 to over 6 times but the second pass kicks into hyper drive. The first pass took over 2 minutes but the second pass took less than one minute.

There's a known bug in the speed detection that isn't fixed until version 2.70.2.4

[Sophoclesdrf]

I suspected that something was awry, but that doesn't explain why I'm getting the speeds to go along with it. The second pass actually takes less than half the time that the first one does, and my encodes are coming in at about 12 to 15 minutes faster. I'll test some more but so far there are some definite gains.

[RaistlinMajere]

Well personally I am not too obsessed with the speed of my computers. It can encode 2 dvd's should I need it to in the time I am asleep so I don't really need it any faster. I have all settings on stable and don't bother overclocking etc. A system that can stay up and running un-rebooted for several weeks is more important to me than outright speed.

The a few things I have noticed which make all results here kind of irrelevant

1. Different films are originally are encoded differently and can give drastically different speeds re-encoding. Different segments of the film also take different amounts of processing. Only comparing using the exact same film can be comparable.

2. Hard drives make a big difference to speed. If you put your working folder on a drive different to your source folder your cpu doesn't have to wait for the drive to keep flipping between the source and the working folder. I have seen this decrease the overall encode time by about 25%. Obviously the faster the hard drive in the first place the better. Also the way cce works at writing files drives can get very fragmented. Hard drives vary results to much to make speeds comparable.

3. Every version of cce ie 2,5, 2.67.x, 2.70,x seems to have a different scale for it speed rating. Therefore only same version results are comparable.

My suggestion is to make sure our systems drives are running optimally, if you haven't got a second hard drive in your system adding one is a cheap way to decrease dvd-rb times. Then see if you can cope with the encode time. If not, upgrade the cpu to what you can afford.

If your already at the point where upgrading processor etc will cost a lot of money ie £300+ just for the processor consider getting a lower spec 2nd pc for encodes, that way your games machines is not tied up and your encoder can take its time.

[on30trainman]

If your already at the point where upgrading processor etc will cost a lot of money ie £300+ just for the processor consider getting a lower spec 2nd pc for encodes, that way your games machines is not tied up and your encoder can take its time.

Exactly what I did. Had an opportunity to take advantage of an offer from Micro Center - $100 rebate on any MB/processor combo. Bought an AMD Sempron 2800+ and Abit MB for less than $100 after rebate. Had extra ATX case, bought some memory, some large HDs and just recently a new video card with DVI output to go with my new Samsung LCD monitor - my old CRT just up and died about a month ago. All video processing programs are on that computer - DVD-RB, Sony Vegas MS 6 and such. Now I can start video processing on that second computer and then switch back to main unit for surfing, tracking this forum, etc. Easy to switch back and forth to keep track of video progress. Actually Sempron unit gives me CCE Basic speeds of about 2.0 to 2.5 in DVD-RB and total processing times of about 2hours or so depending on movie. Very convenient setup.

Steve W.

[gobama05]

I'm pretty happy with my setup. I would love an Athlon X2 cpu, but cannot afford one right now. I get anywhere from 3.1-3.3 depending on the movie I am encoding with my Athlon 64 3500+. I know more DDR memory would help things along. :D

[RaistlinMajere]

@ samuelal about his screenshot above

Can I just point out that your screenshot shows the first segment of a movie which quite often have blank sequences and rarely have any complex action scenes. This obviously gives a far faster encode than you'd expect.

When I used to use tmpgenc I never used to take notice of the estimated encode time until It had got into encoding the main part of the film.

As I said before, without putting some constants in . ie same source, comparing is pointless.

[MedicineMan]

Something must be very wrong in my setup.

I'm using an Atlhon X2 3800; Abit SLI MB; 1Gb RAM; source and work folder in diferent hard-drives; CCE 2.70 (tried 2.67 and 2.50 also); avisynth 2.56; NAN DGDecode; DVD-RB 1.04.

The movie I'm encoding has a reduction rate of about 65%.

And my speed is about 1.5-1.6.


Really don't know what I can do. I've been folowing this thread (and the DVD-RB thread) since the beginning. Don't know if I did something wrong. My previous setup (only different MB with the same chipset and an Athlon 64 3500) was significantly faster.


Hope some of you can help.




Sincerely


MM

[jdobbs]

You definitely need to upgrade to the latest version. There have been four releases since v1.04. There is definitely something wrong. With that system you should be seeing speeds in the 4x range.

You don't have some other application running in the background do you (like the SETI project)? I've seen that slow systems down before.

[RaistlinMajere]

Something must be very wrong in my setup.

I'm using an Atlhon X2 3800; Abit SLI MB; 1Gb RAM; source and work folder in diferent hard-drives; CCE 2.70 (tried 2.67 and 2.50 also); avisynth 2.56; NAN DGDecode; DVD-RB 1.04.

The movie I'm encoding has a reduction rate of about 65%.

And my speed is about 1.5-1.6.


Really don't know what I can do. I've been folowing this thread (and the DVD-RB thread) since the beginning. Don't know if I did something wrong. My previous setup (only different MB with the same chipset and an Athlon 64 3500) was significantly faster.


Hope some of you can help.




Sincerely


MM

The only common complaint with systems that seems to rob all cpu power is when hard drives revert to pio mode. Go to device manager and check to make sure all your ide channels are in the highesst UDMA mode you can select. If your hard drives are serial drives then i am not sure what it could be, check that cce is getting as much of the cpu as it can using the Task Manager

[TotalHavoc]

on my old P4 3.0 O/Ced to 3.5 with a gig of ram I averaged 3-3.4. On my new AMD X2 4400+ (2.2 GHz per core) O/Ced to 2.65 GHz per core I get 5.9-6.2. This is on CCE 2.67

[blueboyec]

Have been using CCE 2.67v (2.4 P4 with .75 gig memo), would there be much of a benefit in upgrading CCE to 2.70.2.4v?

jdobbs,

What version of CCE do you do most of your testing with?

Thanks in advance

[jdobbs]

I do most of my testing with CCE Basic v2.70.1.8.

[MedicineMan]

Thank you for your help.


I still haven't found the reason for the low encoding speed in CCE. At least I found it wasn't a fault in DVD-RB. I encoded a AVI to DVD, and the CCE speed was about 1.1 (I know that there were some avisynth filters in the process, for the resolution conversion).

I checked my hard drives (SATA); defragmented again; checked free space. Found no way to increase the speed.

Going to read the CCE forums to find some help.



Thank you for your time




MM

[jdobbs]

Remember also, that the size and type of the AVI can make a huge difference. For example, if I encode a segmented AVI file that contains DV content - my encode speeds are about half what they would be if reading from a DVD via DGDECODE.

[Boulder]

@MedicineMan: could you check the CPU usage while encoding and report what it says. Did you do a fresh reinstall of the OS when you upgraded your system?

[MedicineMan]

Yes, I did a clean install when i changed my MB and CPU.

I understand that the clock-rate of my new CPU is lower than my previous one (Athlon 64 3500). And i was hoping that the speed would be about the same. I was more or less satisfied until i began following this thread, and seeing the encoding speeds you were getting.

I've tried also CCE 2.50 and 2.67. The speed is about the same. And in Win XP task manager, the performance is above 70 or 80% in both cores. Sorry i can't be more specific about this at this moment. I'll check that.

Maybe i should try a simple encoding, with only CCE. Maybe I could edit the ECL files, take out all except those lines needed to do the encoding, and report back the speed. Trying to isolate the problem. I understand that the result won't (probably) be usable by DVD-RB. But doesn't matter now. I just want to understand what's happening. What do you think about it?


Maybe there's a difference in compression. The DVD I'm using as a test right now needs to be compressed to about 65%. It's a little more than the usual. But is the compression rate so influential in CCE speed?

I'm not at home right now, and so can't be more specific about this. I'll get back with more information as soon as I can. At least I know that the problem lies only in CCE (ok, DGdecode and Avisynth may also be factors here).



Sincerely



MM

[Amnon82]

AMD Athlon 3500+, NF4SLI, 1G 400 DDR >> 2.31x