Slow render times even with more RAM...why?

I have a Quad core Mac Pro that I just upgraded the RAM on from 3G to 7G and can't see any improvement in render speeds, any idea why that would be? As a test I added a simple text box over a single layer of video, added a drop shadow to it (which then requires a render to view) and it didn't take any less time than it did before, about 10 seconds. Shouldn't this have been improved with $225 worth of memory?

If you're talking about editing speed, that's not "rendering." The goal is not to render at all. Really the best things you can do are: have all your material in the same codec, same size, same everything: then you'll have lot fewer of the timeline "bars" indicating rendering. Having a fast enough machine and drives means you can playback in realtime, or closer to it. Many of us are using one of the flavors of ProRes because it's tuned to the editing environment (as opposed to acquisition or distribution). The machines and processors can handle it. Having enough ram means the system does not have to swap out segments of memory to complete operations. As has been noted, FCP can't access all that much memory in the first place. But with too little memory, which doesn't appear to be your case, some of that memory is captured by the OS performing other duties.
I can run ProRes on an older MBP with 4 gigs total using an old Firewire 400 drive without any significant timeline rendering popping up. Of course I'm editing documentary work. I'm not trying to turn it into a Flame for compositing.

Similar Messages

Imovie Export Time Better with More RAM ?

I have an Imac, one before latest release, 3.06 CPU and 2 Gig RAM. A 17 minute video is exporting right now using H.264 and 1920 x 1080 settings and Imovie is telling me that it will take 3 hours. Also, the source video and the exported file are both on an exernal firewire drive (which I know was a mistake).
Question, if I installed 2 gigs more RAM, would it make a big difference ?

I'm a strong advocate for maximizing RAM, but It won't shave that much time of the processing. It still takes a long time to process hd. You will feel the RAM while you work and use multiple applications. It will take out the lag. However, becareful on choosing the RAM. A friend of mine went a really inexpensive route and wound up with conflicts and problems. Its not worth the dollar savings...

What will I notice with more RAM? or FCE HD?

Right now FCE2 seems to be laboring between tasks. Since I just upgraded to the intel it is kind of a disappointment because my g4 Ibook never blinked (at all)... I like the new mac but seems like the old one handled more with less processor and same ram... Is it the older FCE2 or do I need more RAM? Would like to get dialed in for performance... so I can get more work done...

FCE2 is running under Rosetta. Since it is not a Universal Binary program it will not work as well as FCE3. 512 Mb ram is not much even for running FCE3. You would get better performance with more RAM. RAM is very cheap right now, I paid $39.00 US for a 1 GB stick recently. More RAM would help your FCE2 run better. You would also find it improves a lot of other things on your Mac.

Fan on all the time even with apps closed

why is fan on all the time even with apps closed?

try an SMC reset: http://support.apple.com/kb/HT3964

Slow render times with large jpegs - complete system lag

In a project i'm working on I have two large jpegs with a small zoom scaling effect. Going from 100 to 103 percent.
I've noticed that both Adobe Media Encoder and Premiere Pro experience a heavy slow down in render time as soon as the jpegs have to be rendered.
Not only does the render speed almost come to a halt, the complete system lags very heavy, even the mouse cursor won't respond well.
This happens when i have GPU acceleration enabled and when i do a 2 pass H264 encoding.
When I have the GPU acceleration disabled the render goes very smooth, and doesn't seem to slow down...
The jpeg is 4023 x 2677, and 6,97 MB large.
Scaling the jpeg down to about 1920x1080 in Photoshop and put that one in the timeline made the render go a lot faster.
I understand that a large picture takes a bit more time to be rendered, but we're talking about a 10minute render whit the large jpeg file and a 2 minute render with the jpeg resized.
The total time of the two jpegs in the video is 5 seconds in a 3 minutes video.
So, that made me think that the render times are exponentially long.
In the timeline everything runs really smooth.
Is this considered normal, I can't remember having such big differences in CS5. It's not a major thing, but I wanted to share anyway.
My system:
Premiere Pro CC (latest)
i7 4930K
32 GB RAM
2xGTX480
Footage and project on a Raid0 disk
Previews/Cache on a Raid0 disk
System and Premiere on SSD
Render to a single 7200 rpm drive.

>wanted to share
Yes... known issue... I think some of the below is about P-Elements, but the same ideas
Photo Scaling for Video http://forums.adobe.com/thread/450798
-HiRes Pictures to DVD http://forums.adobe.com/thread/1187937?tstart=0
-PPro Crash http://forums.adobe.com/thread/879967

Help - Getting slower render times with AE CS6

Hi everyone
Wonder if anyone else is getting as described?
I have a 3m37s project which is predomnantly motion graphics using live shot footage (.MXF files), Illustrator and a few JPGs.
In CS5 AE I get render times which average around 45mins, so I thought I'd see how quickly CS6 could crank it out by - as you can see I'm getting times which are in excess of 2 almost 3hours!
The project contains a few 2.5D moves as well as tiny bit of Trapcode 3D Stroke
I have mentioned on this forum that I'm having the Error 5070 problems with start up and Ray Trace is unavailable but these times seem seriously wrong to me.
Mac Pro 3,1 (2x 2.8GHZ)
20GB RAM
OS 10.7.4
NVIDIA GeForce GT8800
NVIDIA Quadro 4000 both on GPU Driver 207.00.00.f06
CUDA Driver 4.2.10
All files are on a 2TB drive (7200rpm)
Rendering to a 1TB drive (7200rpm)
Corsair SSD 60gb Cache drive
As an observation when I watch the frames counter ticking over, CS5 seems to steadily work it's way through the render at around less than a frame a second, CS6 seems to crank out 2-6 frames then hold for 30secs before working on another batch. It crawls to a halt near the end.
Can anyone offer any help or advice?
So far I'm not having a great time with my CS6 transition
Thanks
Rob
Message was edited by: Bokeh Creative Ltd
because of a Typo

Thanks Rick - Yes what confused me was that it only took 45mins in CS5 even with MP 'on'
Still having no joy with Ray Tracing though, even though I have a Quadro 4000 card, I get the 5070 error on start up. Any ideas?

Slow render times with latest update

Since updating to Premiere CC 7.1 I am experiencing extremly slow render speed with Magic Bullet Looks and .Filmconvert pro 2. When hitting enter to render it hangs for aout 10 seconds before starting. Then after each shot has been rendered it hangs again before continuing, What used to take around 20 minutes now takes 2 hours 34 minutes.
Adobe, please can you show me how to roll back to version 7.0.1 as i don't have time to mess about with this.
OS: Windows 7
GTX 570 with latest driver (331.65)

Hi captain,
Check out the plug-ins manufacturer's web page and see if there are any updates for your plug-ins.
Thanks,
Kevin

Render time Issues with 5.0 project in CC 12.1

This is actually a copy of a post I had in a different discussion entitled "render time" but this now
I did some extensive testing to try and see why my render times in CC were so much slower that 5.5 and I found some very interesting results.
The original project was created in 5.0 and opened in 5.5 and rendered, it contained video footage and graphics. When it was render in 5.5 it rendered in 4:27. When I opened it in 12.1 and rendered it was 18:00. I then started turning off layers and my render times shot up. basically layer 12 had two animated masks on colored solid. With this off it took 8:30
I then imported the project into 12.1 and saved it, reopened it and with layer 12 off it took 5:11. So importing and saving the project made a difference. Now I recreated layer 12 in 12.1 from scratch and turned it on and my rendered time was 6:50.
So there is something going on with projects being converted from earlier versions - Now to really test this theory I would need to recreate the project from scratch and compare to the 5.5 times.
To further test this we did the same sort of testing with a different comp and found similar results - the render times were different if it was imported vs. just opened. But the big thing we found with this comp was a layer where we were moving large stills with some z space. It looks like the 3d from 5.0 was causing a major slow down in CC on layers that has AE 3d applied.
I think the major take away is if your comp was created in 5.0 and you need to modify it, then make changes in 5.5 or 5.0 or be prepared to recreate any layers with more than simple key frames.
Just to also answer some other questions - in rendering to prores vs. animation or dvcprohd the times were almost the same. The animation was slightly longer but that makes sense cause the file was much bigger.
I also found that with 32 gigs of ram my magic number was 11 reserved for other apps ...
Hope someone can jump in and add some light to this
This was all done on a 2010 mac pro 2.66 w 12 cores 36 megs ram running 10.8.4 - I have 5.5 and CC and the 12. 1 updates are installed and the video card is an NVidia 575. Mulit threading was on and in 5.5 - 9 Megs is reserved and in 12.1 - 10 is reserved

I've experienced similar problem with Java Web Start twice.
Also using JRE version 1.5.0_06 Java HotSpot(TM) Client VM and Browser Internet Explorer 6.0.2800.1106 on Windows XP (Problem was the same with Java 1.4.2)
My application was signed and launched via JNLP. The security warning popup was shown in the windows taskbar, but it was hidden. Using maximize in the taskmanager, I was able to see the dialog but it was blank. The problem was persistent, although I did not try as many times as 20-30 :-)
On both occations the problem occured on multi (3) screen systems, which I suspected to be the problem (Java has historically had some issues regarding multi screen systems), but now I'm not so sure.
Did you find a solution / cause?

Can you start Illustrator CC2014 with more RAM?

I am using Illustrator to edit very large SVG files (>150MB), and the software is extremely slow and stops responding periodically, while it (slowly, slowly) acquires more RAM. Is there a way to start Illustrator with a larger amount of RAM, e.g. 8 or 16 GB?
I'm working on a MacPro with 32 GB of RAM, two 6-core processors, an NVIDIA Quadro K5000 video card (4GB RAM) and tons of free disc space. It's very frustrating to see Illustrator not taking advantage of the available machine resources! I scoured the web for solutions and came up empty-handed.... Any tips?

There is nothing you can do. Even if you could assign a fixed memory range, the program would still need to load the objects one by one and render them, so it actualyl doesn't even make sense to think of it that way...
Mylenium

Slow render time, can anyone help?

I have recently installed some plugins and since it seems that render times are really slow. After every tweak of a filter I'm having to re-render for playback.
I have a one minute clip in the timeline for example, I add a vignette from my Magicbullet Plugin, nothing more and it takes 4 mins to render! My timeline is set to DV-PAL Anamorphic and my video captured as the same.
Any more info required please ask
Many thanks

Your RT settings don't improve final render times, but they may improve your editing experience by allowing better playback of effects without having to render your timeline. You should experiment with the settings in the sequence to see what works best for your particular setting. Generally, if you choose Unlimited RT you'll get more real time playback, and therefore won't have to render as often while editing. But you'll likely experience more dropped frames during your sequence playback. You still have to render your final output though.
Have a read of the manual and experiment. There are no hard and fast rules for settings as every system is different. Obviously, the more you tax your system the less it's able to play back in real time.
Best of luck.

All my TV shows in iCloud on my iTunes appear multiple times and with wrong art, why?

I keep all my past TV show seasons on iCloud, but I don't know why now they appear multiple times (even 10 times) and with all the "art" in Shows tab completely wrong in iTunes.
For instance, Glee in the screen attached appears 10 times: 4 of them with Futurama artwork, 5 with Glee artwork, and once with The Good Wife artwork.
If you see all the TV shows with multiple seasons (Glee, Futurama, The Good Wife) in the screenshot look all messed up and multiple times... this happens with every TV show that I have purchased more than one season... all of it is in iCloud, I keep nothing in my local HDD's.
Since I updated to iTunes 11, this happens.... anyone having troubles with this view????

add your release date manually if it isn't there.
unfortunatly iTunes won't let you do it. i use a program called "TvTagger". just search google, third result from top, that's the one. you have to enter a release date like this: YYYY-mm-dd . I did it and I think it's worth the hassle. not only are your shows sorted correct, but the ugly "Unknown Release Date" line is replaced by the actual release date

Slow render times in Premiere when linking After Effects media

I'm an Avid Media Composer editor but I just switched to Premiere for a special project that needs to have non-interlaced output. I've been pretty impressed until I integrated an After Effects project into my timeline. My render times have now increased exponentially, despite the fact that the After Effects effects are quite simple - single layers of video with very simple moves. Can anyone else chime in with their experience integrating After Effects projects into Premiere? This just seems too horrible to be true.

I've found that the time to render out of After Effects versus the time to render a Dynamic Link comp in PPro is roughly identical. In fact, in many cases it's a bit quicker.
For heavy duty comps, I'd expect to see PPro struggle a little bit more, but definitely not with the basic work you seem to be talking about.

Period account closing bal includes cumul movements, even with a rule.WHY?!

Hello chaps, I've been struggling with this for a few days now and I'm going to have admit defeat. :(
I'm trying to do two things.
1. View periodic balance sheet movements
2. Get the correct custom closing balance to roll up once I've done 1 (in Periodic view)
What I've done.
For Point 1. I use the SwitchTypeForFlow attribute on the flow members. This let's me see the balance sheet movements in Periodic view. Great.
Which leads me onto Point 2. Whilst I can now see the balance sheet movements in Periodic view, the closing balance (the parent of Opening + increases - decreases) does not total correctly. Instead of adding the Opening + increases - decreases for ONE period, It adds them for ALL the previous periods.
Ok, so to get round that, I thought I'd just create a new member called "ClosingBal" and override what HFM is doing. I'll now have two rules. The "OpeningBal" is just the closing balance of last period. "ClosingBal" is just the sum of opening + increases - decreases for the period. So here are the two simple rules
HS.Exp "C1#OpeningBal=C1#ClosingBal.P#Prior"
HS.Exp "C1#ClosingBal=C1#OpeningBal+C1#increases-C1#decreases"
But when I do this, even with the rule, it STILL adds up the cumulative increases and decreases for all prior periods as well, not just the ones for the period. Why is it doing that? All I want is my closing balance in that one period to be equal to opening balance + increases - decreases. It works fine in YTD view, but not Periodic view. I need it to work in Periodic view
any help much appreciated.
Thanks,
Sal

Sal,
There is no solution here that allows the user to simpley change the View member from Periodic to YTD, keeping all else the same, and get your desired result. However, you could create a solution that when a user wants Periodic numbers he/she would select the Periodic View member and then select a unique ClosingBal member (eg. 'ClosingBalPer') and a unique OpenBal member (eg. 'OpenBalPeriodic'). When the user wants YTD numbers he/she would select the YTD View member and then select a different and also unique ClosingBal member (eg. 'ClosingBalYTD') and a different and also unique OpenBal member (eg. 'OpenBalYTD').
Here you would have 2 different rules for you opening balances:
HS.Exp "C1#OpeningBalPer=C1#ClosingBal.P#Prior"
HS.Exp "C1#OpeningBalYTD=C1#ClosingBal.P#Last.Y#Prior"
And your ClosingBal rules would look like this:
HS.Exp "C1#ClosingBalPer=C1#OpeningBalPer+C1#increases-C1#decreases"
HS.Exp "C1#ClosingBalYTD=C1#OpeningBalYTD+C1#increases-C1#decreases"
In regards to the ClosingBal rules, these aren't necessary if you make the ClosingBal member a parent of the other members. If 'decreases' need to be subtracted then they should be flagged as 'SwithSignForFlow'.
With this solution if a user selected YTD with the Periodic members or Periodic with the YTD members then their numbers wouldn't make sense so there would be some user eduction involved with this solution.

Slow render time in Premiere Pro CS4 (with fast computer?)

Hello people on the Adobe Forums,
I hope someone can help me with this question, it's been bothering me a long long time now and this week i have to make a decision, or fix the problem and keep this configuration, or spend money for a new cpu (i5 750) and a new motherboard. But i think this computer should do better then it does. I have the following configuration:
AMD Athlon X2 4600+
Nvidia GeForce 7800GTX
2× 1Gb OCZ memory (DDR, PC2400, 250mhz)
Storage Disks:
Seagate 320GB SATA (7200, i think)
WD 74GB SATA (10.000RPM, OS)
Hitachi 500GB SATA (also 7200, i think)
WD 1TB USB (SATA) (extern hdd)
WD 250GB SATA (7200, i think)
So, i don't understand the problem, Premiere is running very slow on my computer.. The rendering of a 5min piece of just a random movie takes like 3 hours! Just right now i am rendering some things I filmed with my mobile, it does like 1 frame in 10sec!! On most pc's with this vice versa!
So the cpu is not that old, and neither is the videocard, meaby the memory is a little bit low.. Could that be the problem?
Hope you people can help me out, i searched Google and everywhere but can't find anyone with this problem..
List of software:
- Windows 7 Ultimate
- Adobe Premiere Pro CS4
Drivers are, as far as i know, all up to date, i updated the Adobe programs too..
Describe source material, Codec: XDCAM-EX Movie (mp4)
But like i said, it happens too avi too, it think that was DivX..
I bought this pc second-hand, so it could be that something in the processor is already broken and i didn't notice it at all.. i searched like everywhere and also tried too reinstall the program but that doesn't work either..
Some more info:
http://www.creadion.nl/premiere/info.jpg
http://www.creadion.nl/premiere/info.nfo
http://www.creadion.nl/premiere/DxDiag.txt
Ohye, and the PPBM: http://www.creadion.nl/premiere/Output_PPBM4.txt
With friendly regards,
Marinus

Well, I'm helping other people too here on some dutch foras and i hate how people say things like "Doesn't work. Help" and just wait and wait for someone to answer. Plus i want to order my new hardware this evening (10:27pm now here), so meaby that's a 'cause' too, to make you people answer fast! (and it worked! hehe)
So, what i am thinking of right now is to upgrade the motherboard, cpu and ram.. The videocard can wait a little more:
- Motherboard: MSI P55-GD65 (not sure of this one, good reviews&tests, Asrock P55 Extreme is an option too.. be my guest to do another recommendation) (Asus P7P55D seems like a good option)
- CPU: Probably an I5 750 from Intel, heard good things about this one.. Although i've always used AMD..
- RAM: I really don't know, don't see many differences between the RAM sticks.. But yeh: OCZ Gold OCZ3G2000LV4GK
Sorry for no links to product pages, you have to do a little search, i only got some dutch websites, and i know for sure that you all having you're own comparing/specs websites (which i don't know) .
Thank you very much untill now, meaby i get another reply tonight.. Ohye, plans changed, i'm not ordering tonight anymore.. I'll do that tomorrow morning (I'll do a few calls to a view webshops and see if they got in stock, if so, i'm gonna drive (or go with bus) to one of them.. I need it really really quick, need to submit a project from school August 30
Thanks everyone!
Just took a look at the links you gave me, the Asus P7P55D is an option too!
Took a look at the links, added the "Asus P7P55D".
What i forgot to mention: I got a budget with a max of like €400,- (That's like 510usd, I don't even really have the money) so I can't buy a whole new pc..

Serial VISA 'Write' -why is it slow to return even with large buffer?

Hi,
I'm writing a serial data transfer code 'module' that will run 'in the background' on a cRIO-9014. I'm a bit perplexed about how VISA write in particular seems to work.
What I'm seeing is that the VISA Write takes about 177ms to 'return' from a 4096 byte write, even though my write buffer has been set to >> 4096.
My expectation would be that the write completes near instantly as long as the VISA driver available buffer space is greater than the bytes waiting to be written, and that the write function would only 'slow down' up to the defined VISA timeout value if there was no room in the buffer.
As such, I thought it would be possible to 'pre-load' the transmit buffer at a high rate, then, by careful selection of the time-out value relative to the baud rate, it would self-throttle once the buffer fills up?
Based on my testing this is not the case, which leaves me wondering:
a) If you try to set the transmit buffer to an unsupported value, will you get an error?
b) Assuming 'yes' to a, what the heck is the purpose of the serial write buffer? I see no difference running with serial buffer size == data chunk size and serial buffer size >> data chunk size??
QFang
CLD LabVIEW 7.1 to 2013

Hi, I can quickly show the low-level part as a png. It's a sub-vi for transferring file segments. Some things like the thin 'in-line' VI with (s) as the icon were added to help me look at were the hold-up is. I cropped the image to make it more readable, the cut-off left and right side is just the input and output clusters.
In a nut-shell, the VISA Write takes as much time to 'return' as it would take to transfer x bytes over y baud rate. In other words, even though there is suppused to be a (software or hardware) write and read buffer on the com port, the VISA write function seems to block until the message has physically left the port (OR it writes TO the buffer at the same speed the buffer writes out of the port). This is very unexpected to me, and what prompted me to ask about what the point is of the write buffer in the first place? -The observations are on a 9014 RT target built in serial port. Not sure if the same is observed on other targets or other OS's. [edit: and the observation holds even if transmitting block-sizes of say 4096 with a buffer size of 4096 or 2*4096 or 10 * 4096 etc. I also tried smaller block sizes and larger block sizes with larger still buffers. I was able to verify that the buffer re-size function does error out if I give it an insane input buffer size request, so I'm taking that to mean that when I assign e.g. a 4MiB buffer space with no error, the write buffer actually IS 4MiB, but I have not found a property to read back what the HW buffer is, so all I have to base that on is the lack of an error during buffer size setting.) [\edit\]
The rest of the code is somewhat irrelelvant to this discussion, however, to better understand it, the idea is that the remote side of the connection will request various things, including a file. The remote side can request a file as a stream of messages each of size 'Block Size (bytes)', or it can request a particular block (for handling e.g. re-transmission if file MD5 checksum does not match). The other main reason for doing block transfers is that VISA Write hogs a substantial ammount of CPU, so if you were to attempt to write e.g. a 4MiB file out the serial port, assuming your VISA time-out is sufficiently long for that size transfer, the write would succeed, but you would see ~50% CPU from this one thread alone and (depending on baud rates) it could remain at that level for a verrry long time. So, by transferring smaller segments at a time, I can arbitrarily insert delays between segments to let the CPU sleep (at the expense of longer transfer times). The first inner case shown that opens the file only runs for new transfers, the open file ref is kept on a shift register in the calling VI. The 'get file offset' function after the read was just something I was looking at during (continued) development, and not required for the functionality that I'm describing.
QFang
CLD LabVIEW 7.1 to 2013

Slow render times even with more RAM...why?

Similar Messages

Maybe you are looking for