Speeding up rendering.

I am editing some HDV footage of an interview, and render speed it annoyingly slow, and I had two questions about how to speed up the speed.
I realize this is a stupid question but it is something i have never really thought about until now. I am overlaying other shots with the interview footage underneath.How exactly does FCP handles multiple layers and would it would speed things up to cut out the interview footage underneath in the sections when it is just being overlayed with other footage?
Second question is I was looking at render control, and I was wondering if changing the resolution would speed up rendering. When played around with it, it seemed like it didn't change render times.
Thanks
-JCP

jcompagni wrote:
Thanks for all the advice but I was really just looking for answers to my two questions from the original post.
Your OP actually contains more than two questions and some lovely openings:
I realize this is a stupid question <</div>
You'll notice we didn't really use that for material.
but it is something i have never really thought about until now. < </div>
Umm, might be a good idea.
How exactly does FCP handles multiple layers and would it would speed things up to cut out the
interview footage underneath in the sections when it is just being overlayed with other footage?< </div>
FCP does not handle multiple layers at all, it just plays the top most layer. No, it would not speed anything up to do whatever it is you think you could be doing.
Second question < </div>
Nope, this is question number three.
is I was looking at render control, and I was wondering if changing the resolution would speed up
rendering. When played around with it, it seemed like it didn't change render times. < </div>
If you empirical knowledge indicates one thing, why do you feel you need us to confirm your research? Your issue isn't about rendering, it's about using HDV, which is not an appropriate editing format. You may be able to play it back in real time on your powerful Mac but playing HDV back as a streaming decompression is different from rendering which requires individual frames to be created, one at a time, from your long-GOP, compressed HDV format.
I hope we have now answered all of your questions.
bogiesan

Similar Messages

Can i hack gtx 560 ti speed up rendering in premiere pro using mercucy engine ??*

I'm going to buy this card soon, msi gtx 560 ti oc version, i wonder if i can "hack" it to use mercucy to speed up rendering and playback ??? please tell me ??

As Harm stated, yes, you can. However, there are currently two different MSI GTX 560 Ti OC cards on the market (the Twin Frozr II and the Twin Frozr III, the former being based on the regular 560 Ti with only 384 CUDA cores and 256-bit graphics memory bus, the latter with 448 CUDA cores and a 320-bit graphics memory bus). Between the two, go for the Twin Frozr III as the lesser Twin Frozr II (being based on the "regular" 560 Ti) can only achieve a score of 90 to 100 seconds (versus 60 to 70 seconds for the Twin Frozr III/GTX 560 Ti 448) in the MPEG-2 DVD portion of the PPBM5 benchmark even on a system with a heavily overclocked CPU. (This assumes that you're going to use the card in a system with an overclocked i7-2600K or higher CPU. If on the other hand you're going to use the new card in a system with only a limited-overclockable i5 CPU or an i7 CPU that's stuck at or near stock speed, then the regular Twin Frozr II would be sufficient as the Twin Frozr III's performance would not have been sufficiently faster in such a limited system to justify the higher price.)
Remember, performance is affected not only by the number of CUDA cores, but also by the memory throughput. The latter is the very reason why the GTX 550 Ti is slightly slower than the GTX 260 despite both GPUs having the exact same number of CUDA cores (192, in this case) because 192-bit DDR5 memory (as used on the GTX 550 Ti) has slightly less total memory throughput than the GTX 260's 448-bit DDR3 memory.

Is their a 3rd party card that will speed up rendering for FCP6?

Is their a 3rd party card that will speed up rendering of DV footage on the timeline & for conversions also for FCP6? I am thinking about buying an 8 core.
steve

8 gigs, one per processor.
But you're not going to achieve any appreciable decrease in rendering times.
It's a topic that we see around here about one a week so you can search for RENDERING TIME or similar topics and see lots of threads.
If you want faster rendering times, you have to use a proprietary, hardware-based codec. This is NOT a good idea unless you are prepared to abandon the hardware when the project wraps since it will certainly and quickly become obsolete or orphaned when the mfr goes out of business.
bogiesan

How to speed up rendering/ transcoding in FCPX

It has just taken 8 (eight) hours to render then transcode two hours of video footage, before any editing.
I have two video clips from a Panasonic camera in .MTS form. (1920 x 1080)
To get them onto FCPX I used Handbrake to convert them to .mp4 files.
On FCPX import preferences, 4 items ticked, Copy file to event, Create opti media, Create proxy media, Analyze audio.
I also then clicked stabilization, on both clips.
Machine is a Mac Pro serial # YM******XYK with upgraded video card and 20g RAM (4x 4G +4x 1G)
Using OS 10.6.8
Using USB hard-drive to store original footage, but
onboard hard-drive for event and project
Activity monitor showed most of the time CPU's at 30% use 60% idle and RAM 7.7M used. 13M idle.
There was a short period when FCPX CPU at 750%, however RAM never went above 8M
I know best way to quicken render / transcode is buy a bigger computer,
However would like to know if I can get better % use out of CPU and RAM
It would appear that I did not need any more RAM above 8M, as machine appears not to use rest.
Also, I would like to know if I am bringing footage into FCPX, in the most efficient way.
PS, footage is of a hall presentation. Sound taken off camera mic and camera operator keeps moving / bumping camera. So need to analyze, then optimize both audio and video to make best of a bad shoot.
<Edited by Host>

A couple of points that may work.
Rendering -
     this is only used so you can see the finished 'look' while working on your project. If this is not important, turn rendering off in Preferences. This appears not to affect final output, but quickens work time.
Original files V's Proxy media.
     Working with Proxy media speeds work times and the computer processing. So use Proxy unless the original files needed.
To do this you need all projects in FCPX closed.
In preferences -
          Import, tick 'Create Proxy media'
          Playback, tick "Use Proxy media'
This was for version 10.0.3
Note if you change these preferences while you have an existing project, you can cause files to go offline.
Only a novice, so if someone who knows better, they can expound a better way to speed up the computer processing

Speeding up Rendering in LR3: Faster Processor or Better Video Card?

I'm currently running a MacPro with 2 Dual-Core Xeons at 2.66 GHz and 7 GB of RAM. I upgraded a while ago to an ATI Radeon HD 3870 video card.
I'm a professional photographer who is wondering what would give me the best LR3 rendering speed bump for my buck. Should I...
1. Add more RAM?
2. Upgrade my Video Card? (If so, which one?)
3. Start from scratch with a faster Mac?
4. All of the above?
I'm looking for any information that would show me real world measurements using LR3. I'm NOT looking for philosophical discussions or speculation. Has anyone seen a test that I could read, or information source that compares different systems rendering identical content in LR3? Concrete help (rather than speculation) would be appreciated.
Thanks!

ambienttroutmask wrote:
Good hard fast hard drives can also help, cheap and can carry on to any new machine, so always worth sticking one in. The Samsung F3 HD103SJ 1TB is an absolute bargain and for real world speed only SD is faster and then only if the OS is on the SD. I got one recently for video editing and stuck my LR catalog on it, a definite noticeable speed increase over other hard drives of a similar spec (although I have heard nasty stuff about them failing early...so back up obviously. I also have a big fast video card for video editing, that sits there doing absolutely nothing when using LR...so unless you edit video or like playing games, don't bother. I prefer my old machines ancient Matrox card for stability and 2 D rendering, but it won't do video sadly!
Processor comparisons here. With your set up you are going to have to invest in a top of the range MAC with an i 7 to benefit anything significant. More RAM won't hurt but may not get used by LR. More RAM and a i7 processor will ensure the RAM actually gets used.
Pete,
Just curious. Have you done any measurements on the Samsung? Say with SpeedTools for the Mac? I'm just wondering what the read/write speeds are across eSATA connection vs. any of the others.. I know a lot can vary from one machine to another, but I was just asking on your machine how it compared with a measured result to another brand. Like you say, the specs are very similar to other drives.
Thanks.
Jay S.

Speeding up Rendering during production

We are using FCP to produce a product called www.GloryScapes.com. Our problem is the rendering time. We are creating overlaid text in Motion and importing it into FCP and we have to wait for these words to render with the video footage an inordinate period of time. We have tried setting FCP to "draft quality" in the "Render Control" so this will move along faster during the production process, then we can change it to "best" once we have completed the program. Any other suggestions to speed up the rendering?

That was a great suggestion. Yes, the settings match that of FCP. We are working with DVCPRO HD 1080i60 (1280x1080) 16:9. We'd be welcome to any other suggestions. Thanks.

What can I do to speed up rendering?

THIS QUESTION. HAS. BEEN. ASKED. TO. DEATH. I KNOW
But the answers have become so common that someone like me who knows how to use AE but doesn't know about rendering does not understand the jargon
I rendered a clip using Trapcode Particular and some flux to create a particle trail, and behind that I had a clip of paticular running with some glow. This is it.
http://www.youtube.com/watch?v=fedcUIb1OIo
NOW, that render took about maybe 5 minutes? probably less I don't remember, but it was short.
I have renedered large clips in Pr and it takes at max maybe 15 - 30 minutes with tons of effects(short of warp stabalizer, and with every single clip having some type of Magic Bullet Looks effect). That took maybe about an hour for a 4 minute video.
I go into AE, and I create this. https://www.mediafire.com/?iqg25p7i62826iw
If you have Trapdcode, it should open, but since that's just an idea of what I have, hopefully I can explain whats in it.
I have one comp where there's 3D text that has ray traced 3D and been moved to have a kind of corner alignment
The next comp is THAT SAME ONE I JUST POSTED I just shifted it into this project/comp and took out all the backgrounds and other particles, now the only thing there is the "flame" particles. I also tracked that one to move throughout the first 30 seconds and then disappear
Next comp is an audio spectrum, duplicated and fliped so the audio at both ends is the starting frequency.
I precomped a lot of ****, but I cleared out what I didn't need and only have two comps left in the main composition.
Why is it this render is taking 28 hours? Nothing on earth ever took this long for me, and you're talking to the guy who warp stabalized a 2 hour film because he could.(EVEN THAT TOOK 6 HOURS, UNDERSTANDABLE)
What can I do to somehow reduce this time? I've already done the render as a quick time format thing, that reduced it to about 11 hours. But this clip is only 1:35 long.
whyyyyy
System Information
Time of this report: 1/2/2014, 23:23:01
       Machine name: MGREENSCOMPUTER
   Operating System: Windows 7 Home Premium 64-bit (6.1, Build 7601) Service Pack 1 (7601.win7sp1_gdr.130828-1532)
           Language: English (Regional Setting: English)
System Manufacturer: Dell Inc.
       System Model: XPS 8300
               BIOS: BIOS Date: 02/15/11 09:36:59 Ver: 04.06.04
          Processor: Intel(R) Core(TM) i5-2300 CPU @ 2.80GHz (4 CPUs), ~2.8GHz
             Memory: 6144MB RAM
Available OS Memory: 6126MB RAM
          Page File: 4724MB used, 7526MB available
        Windows Dir: C:\Windows
    DirectX Version: DirectX 11
DX Setup Parameters: Not found
   User DPI Setting: Using System DPI
System DPI Setting: 96 DPI (100 percent)
    DWM DPI Scaling: Disabled
     DxDiag Version: 6.01.7601.17514 64bit Unicode
DxDiag Previously: Crashed in DirectShow (stage 1). Re-running DxDiag with "dontskip" command line parameter or choosing not to bypass information gathering when prompted might result in DxDiag successfully obtaining this information
DxDiag Notes
      Display Tab 1: No problems found.
        Sound Tab 1: No problems found.
        Sound Tab 2: No problems found.
          Input Tab: No problems found.
DirectX Debug Levels
Direct3D:    0/4 (retail)
DirectDraw: 0/4 (retail)
DirectInput: 0/5 (retail)
DirectMusic: 0/5 (retail)
DirectPlay: 0/9 (retail)
DirectSound: 0/5 (retail)
DirectShow: 0/6 (retail)
Display Devices
          Card name: AMD Radeon HD 6700 Series
       Manufacturer: Advanced Micro Devices, Inc.
          Chip type: AMD Radeon Graphics Processor (0x68BA)
           DAC type: Internal DAC(400MHz)
         Device Key: Enum\PCI\VEN_1002&DEV_68BA&SUBSYS_31501682&REV_00
     Display Memory: 3823 MB
   Dedicated Memory: 1015 MB
      Shared Memory: 2807 MB
       Current Mode: 1440 x 900 (32 bit) (60Hz)
       Monitor Name: Dell SE198WFP(Analog)
      Monitor Model: DELL SE198WFP
         Monitor Id: DELF003
        Native Mode: 1440 x 900(p) (59.887Hz)
        Output Type: HD15
        Driver Name: aticfx64.dll,aticfx64.dll,aticfx64.dll,aticfx32,aticfx32,aticfx32,atiumd64.dll,atidxx64.d ll,atidxx64.dll,atiumdag,atidxx32,atidxx32,atiumdva,atiumd6a.cap,atitmm64.dll
Driver File Version: 8.17.0010.1191 (English)
     Driver Version: 12.104.0.0
        DDI Version: 11
       Driver Model: WDDM 1.1
Driver Attributes: Final Retail
   Driver Date/Size: 3/28/2013 21:37:02, 1155264 bytes
        WHQL Logo'd: Yes
    WHQL Date Stamp:
Device Identifier: {D7B71EE2-2BFA-11CF-9571-5A11BEC2C535}
          Vendor ID: 0x1002
          Device ID: 0x68BA
          SubSys ID: 0x31501682
        Revision ID: 0x0000

Yeah, you get what you asked for - you used one of the worst 3D raytracers on this planet without GPU acceleration and shutting it off makes all the difference. Really nothing unusual here. That is why people use plug-ins like Element 3D or Invigorator or in CC Cineware...
Mylenium

Looking for a way to speed up rendering when only changing a JPG in HD Video

Hey guys,
Wondering if anyone can help. I have a movie that I render on a weekly basis for a business which has various HD video clips as well as a background JPG layer. Fundamentally, the video clip plays for 30 seconds followed by 30 seconds of the JPG, and this alternates for about 15 minutes. Every week I have to change the JPG to reflect what is current for that week, which means I have to re-render the entire project which is pretty time consuming. Is anyone familiar with a way to speed up this process? All help is greatly appreciated.

Perhaps you misunderstood. If I understand your process correctly, you have one video over one jpeg background that repeats and repeats.
My recommendation: don't render the same thing over and over again in AE. Render in AE ONCE, then take it into the editing app to repeat and repeat.

Speeding up my rendering

Okay here's my question,
is the speed of rendering directly related to the size of your video card. The video card I have installed is a radeon HD for 4800 I believe it has 1 GB of RAM, but I don't know if megahertz my rendering seems to be slow. Especially the more complicated. The graphics applied, what would be the best recommendation to speed up my rendering. I'm also running a i7 processor, quad core with 12 GB of RAM. I know it's overkill that someday they'll be writing that big. Thank you in advance

You don't mention which version of PP you're running.
For CS4 and earlier, the best way to speed up rendering/exporting is with a faster CPU. Nothing else will give you the same degree of performance improvement as that will (assuming nothing else in the system is the bottleneck).
For CS5 or later, you can also speed up rendering/exporting by adding an appropriate nVidia graphics card, which is capable of accelerating some aspects of the job. (Sorry, ATI won't work for this, it has to be nVidia.)

ATI 8800 vs. Quadro 4000 / 5000 Rendering Speed

Fellow Forum Members,
I'm building a system to run Premiere Pro at an optimum level. However, the conflict I'm encountering relates to Quad Monitor Support. Every single CUDA card offered by Nvidia (i.e., GTX 580, Quadro 4000, 5000, 6000) seem to only offer Dual monitor support. The third mini HDMI port is worthless since it doesn't run together with the two DVI ports. Not happy with Nvidia about this decision.
I'm building a Windows 7 64 bit system and It's my understanding that if I want to run Premiere Pro CS5 using the Mercury engine to speed up rendering, then the only game in town is an Nvidia Quadro graphics card. I wouldn't have a problem with this only if Nvidia would offer a video card with Quad Monitor support like ATI does with the 8800 card they sell.
So my question is, "Does Premiere Pro render at the same speed on an ATI 8800 card vs. a Quadro 4000/5000?" If the answer is yes, then I'm buying the ATI 8800 since this card supports 4 monitors. If the answer is that the ATI 8800 is slightly slower with rendering, then I'm still buying an ATI 8800 since a slight difference in rendering doesn't matter that much to me. However, if the answer is that Premiere Pro rendering on an ATI 8800 is serioiusly handicapped. Then I have a problem, since I really want Quad monitor support and Premiere Pro to render fast.
I hope to hear back from Premiere Pro users that have used Premiere Pro on platforms equipped with ATI 8800 and / or Nvidia 4000, or 5000. Did you notice any performance change between ATI 8800 and Quadros? Thanks in advance.

icegene wrote:
Hi there,
I was wondering if you have found an alternative solution to buying two quadros. I once came across a post in the nvidia forums about putting a Geforce Card and a Quadro in the same machine...
Also, I've seen that GeForces don't need to be in SLI to give you more than 2 monitors but you will need more than one card. Two cards in SLI will only support 2 monitors while a third card can support another 2. giving you 4 ports. As seen on the link, you might be able to pull of up to 6.
Alternatively, there is the Quadro NVS cards that have up to 4 display ports.
Hope this helps!
Heaven forbid SLI and Premiere together, do not even think about it.
Here is my suggestion.
Just for you I tried adding a GTX 550 Ti to my system which has a GTX 580. I had two monitors on my GTX 580 and while I did not have two more spare monitor I did have of additional 1920 x 1200 monitor. I powered up and it found and installed another driver for the GTX 550 Ti, I do not know why it did not install a second copy of the 267.59 driver but installed 266.58. It all worked I then had three surfaces all connected and then I ran our PPBM5 benchmark and it was not affected by the the addition of the second board. By default it used the CUDA of the first board. Of course I cannot guarantee other drivers or other GPU's or for that matter other motherboards. Make sure you have lots of power available I would not mix Quadros and GeForce boards and why would you want to anyway. After all that I suspect that when you add the other realtime software you probably will ruin any editing capability you might of had.
I just took a look at the prices, the two board nVidia solution is about half the cost of your ATI V8800 solution!
Message was edited by: Bill Gehrke

Speed increase during rendering

Hi all.
I made a strange observation that needs verification or trashing.
While working in a project I selected render (as allways) but after this I
clicked out in the Finder = deselected FinalCut. Then to my surprice
the rendering process appeared to speed up.
Is this a reality or just a subjective feeling ?
Yours Bengt W

No that's the problem. I only got the feeling while doing several adjustments
with need for rendering. I let it be every second time (say odd) and selected finder the even time and got this impression/feeling.
That's why I addressed this "observation" to find out if othere run into a similar
experience.
(will keep on doing it the "Finder way" as long as I get the feeling of an increased
speed in rendering)
If this is a real/measurable phenomen then Apple programmer ought to sit
down and take a look on the rendering process and optimice it's coding.
Else: it's just in my mind and of no interest.
Yours Bengt W

The need for Speed - my setup

I have been using Lightroom since version 1 and have recently upgraded to version 3. As many have reported, version 3 needs more CPU speed for rendering so I have modified my workflow to get maximum speed.
Until recently I had a Dell M65 Notebook with a 2.16 GHz Core2 dual CPU and 3 GByte RAM. I was using Windows 7 x64. All my RAW files are on an external HD, connected with USB 2.0 to the notebook.
Upgrading to Lightroom 3 made everything too slow, as the CPU would stay at 100% for long times, like 10 seconds during edits.
Therefore, I upgraded to a Dell M4500 with a Core i7 CPU Q840 @ 1,87 GHz and 8GByte RAM. This CPU has 4 phyical cores, each supporting hyperthreading for a total of 8 logical cores. I am using Windows 7 x64.
On the new notebook, I have never had L2 installed.
The L3 Library is located on the internal Hard drive, together with all the previews. The RAW files are still on the external 2.5" USB 2.0 drive.
I upgraded by importing first the L3beta catalog and then L2 catalogs. I found a bug, which would crash L3 if the same photo was already in my L3 catalog as well as in a L2 catalog to be imported. I solved that problem by importing the L2 catalog first separately and then exporting parts of it to new L3 catalogs and then importing these parts to the main L3 catalog.
During import, I had the external drive holding all the photos attached to the notebook.
At this point, I had only the catalog in place, but no previews. I set the preview size in L3 to 2048pixels. It is important to set this value higher than the screen resolution. Otherwise, L3 will have to process the RAW files most of the time, making things very slow.
I once tried looking at the imported photos during import. That was not a good idea as L3 went very slow and started swapping on the hard drive. I had to stop import and restart L3 to regain speed. I assume that L3 has some memory leak issues.
Now I opened one path on my external drive I wanted to work on. I noticed that when I opened a photo in the Library page, I would see the "Loading". After about 2-3 seconds for old (2003) photos and about 3 seconds for new (2010) photos, it would go away. This would of course not allow me to browse through at high speed. Therefore I selected all photos in one folder with new photos and asked L3 to make all the standard previews. L3 would start making them at a speed of about one every 1-2 seconds. Those photos were without any fancy stuff like adjustment brush. One thing I noticed was that the CPU load was between 25 and 50%, so L3 is not using all the available cores for this operation. Optimizing this process for many cores would give me a speed advantage of 2-3.
I have about 50k pictures in my library. Almost all of them RAW files.
The result is that I can now browse through the photos with a very low delay. If I select the next photo, I will see a low res version of it almost instintaniouosly, and the high res version after about half a second. If I go crazy and use the scroll wheel extensively and very fast, after some 5-10 or so photos, scrolling stops. At that point, I have to wait for the film strip at the bottom of the screen to update. Then I can continue. If I do not show the film strip, I can scroll a bit further, but there are still intermittent stops, where I have to wait for L3 to catch up. Still, even at the point where scrolling stops, I still get the hi res picture after about a second. It seems that the speed of the internal Hard drive is the limit here. If I need to scroll faster, I select the grid of pictures. There I can go even faster.
Zooming in to 1:1 forces the full resolution previews to be made. That takes about 3 seconds on my notebook. If I revisit the same photo later, I can zoom in and get the full res photo in about half a second.
I then tried to import some data from the CF cards. The import went smooth (although rather slow, but that was the same in L3b2) and I tried to look at the pictures during import. That was not a good idea, as L3 started using RAM excessively, the hard drive was working continously and everyting went very slow. I stopped the import and restarted L3. Apparently, there is a memory leak somewhere so I decided to adopt the following workflow:
Import the files for the Memory Card (I have a USB card reader)
Close L3 and restart (I skip the backup if asked by L3)
Select all newly imported pictures and generate the standard previews.
Close L3 and restart (I skip the backup if asked by L3)
Now I select what pictures I want to work on by by giving them the 1-5 stars.
Then I filter the 5 star pictures and move over to the Edit module
If I see the internal Hard drive working a lot during edits, I close L3 and restart.
Browsing in the Edit module is much slower than in the Library module.
Also, during edit, speed of operation is dependent on the resolution I look at. If I look at the photo zoomed out, I get to see the next picture after about 1/3 of a second, and have a waiting time of about 3 seconds the first time I open a picture until the "Loading" goes away. I can only assume that at this point, the 1:1 previews are made. If I go back to a previously edited photo, the time delay is about 1 second. If I close L3 and reopen, it takes 3 seconds again. During that time, all 8 cores are working, but only up to about 50%, so there is still room for improvment here.
Browsing zoomed in takes initially about 1 seconds extra. Returning to a photo zoomed in does not seem to cause any extra delay, though.
I always wait until "Loading" goes away before I try editing anything. If I start immeditaly changing something, L3 becomes slow. Sometimes it stays that way, and I have to restart to regain speed.
The Memory for Camera Raw is set at 1 GB, the default. I have not found any speed advantage setting it higher.
I use the adjustment brush a lot as well as red eye removal and gradients and I see no practical difference in rendering time when used. If I paint on the photo zoomed out, I see the result of the adjustement brush instintaniously during painting. Changes to the exposure etc. is also rendered with no visible delay.
Editing zoomed in to 1:1 is another story. If I use the adjustement brush, I have to wait about 3 seconds for anything tho happen after I made the first stroke. After that, the delay is reduced to about half a second. L3 does record the movement of the mouse correctly, but the update is much slower than if zoomed out. Changing the exposure dial is also rendered with about half a second delay. Therefore I try to do as many edits zoomed out as possible.
Looking at the memory consumption after a couple of hours of work, it seems that L3 needs several GBytes of RAM to be happy. When I close L3, about 3 GBytes are freed.
Hope this can be of some help. Questions and Comments are wellcome.
Best regards,
Bengt

Setting Lightroom's preview size to larger than screen resolution is guaranteed to slow things down. I have a dual monitor system whith Lightroom showing the loupe on one screen and the grid on the other. The screen sizes are 1920x1080 and 2560x1440. Following your practice for previews would bring my system to a crawl and I have an entry level workstation.All my previews are of standard 1024x768 in size. The only delays in viewing come when Lightroom is asked to display an image where there is no preview. Once the preview has been created, browsing is near instantaneous.
I beg to differ! The reason is that a preview can be used for any size, which is smaller than that of the preview. Scaling down a preview can be done with little or not loss in quality compared to getting the full resulution and scale that down. Scaling up is not possible.
In your case, you will not use the previews of 1024x768 at all for full screen display. Instead you will force L3 to create bigger previews the first time you open the photo. You could try this by generating the standard previews for new imported photos before you start browsing. Then try to browse. You will probably get the "Loading" message, which indicates that the preview was not used and another one with higher resolution will be made. Then try the same experiment, setting the size of the standard preview to 2048. If, and only if, the size of your display area is not more than 2048 pixels wide and you display a photo in lanscape orientation, you will be able to browse without delay after generating the standard previews.
If I understand your setup correctly, you display the photos in a window of 2560x1440 pixels. I suspect that only 1:1 previews could be used for that. I think Adobe should increase the maximum size of the previews to a value larger than 2048, so that you are not forced to use only the 1:1 previews.
I agree, that when the previews have been generated, browsing is near instantaneous.

Mercury Engine speed up all types of footage?

I will be working with some native 5D footage, and alot of footage that has already been converted over to pro res.
Does having a card that can utilize the mercury playback engine speed up rendering on all kinds of footage? Or only very high quality/AVCHD?
I have alot of footage that's in photo-jpeg as well. Just wanting to make sure I understand how the system works.

Looking at the GT240 card using the hack

Rendered Video has Glitches

I'm from Toronto, Canada and right now working on an online for a PAL tv series. This project's been a long process, originally all the required files were on the suite's FW800 RAID drives but now they're all on drives supplied by the client: FW 400, 300 GB external drives, no RAID (just FW 400 daisy chain).
I'm onlining using a 10-bit PAL timeline, making sure all my elements are in place, then I render this timeline. Upon playing back the render, there's a few visual glitches every now and then (e.g. flash of a pink frame, or green blotches showing up on text renders). If I re-render these problem sections, the errors go away.
Does anybody have any idea why these errors are appearing in the first place? Maybe the slower read/write of the FW 400 drives are causing a render issue? Yes, I do realize I would be unable to playback my timeline due to FW400's slow speed.

Wouldn't the computer 'slow down' to compensate for the slower drive speed when rendering? So the drives can't keep up with the speed the computer sends the rendered frames back so it only gets 'part' of the data for that frame?
I adjusted the opacity setting for it and the blotches disappeared...
Lower 3rd was a text track, a semi-transparent black rectangle behind it, and finally a DV PAL clip in the background. This was all being rendered in a 10-bit PAL timeline. Effects Rendering? Pretty sure only Final Cut is doing that on this system. Unless an AJA IO would have any effect.

Speed disappointment

I talked my sister into getting a iMac 24" 2.16GHz. (Her first Mac)
After showing her how to use it, I decided to do a test to see how much speed I would gain if I got one.
I brought along a copy of an iMovie/iDVD project on a FW disk and copied the whole thing to the HD on her new Mac.
I then tested the speed of rendering a 2 hour video (36 gig iMovie project) which my Mac takes around 6 hours to compress and encode before it begins to burn the DVD.
I was looking for a HUGE speed boost that would convince me to shell out the $2000.00 to get one for myself.
Using her computer, I would only save about 30 minutes.
I guess for $500.00 more I could get a MacPro which has Two 2.66GHz Dual-Core Intel Xeon “Woodcrest” processors, but would I see any REAL increase in speed?

.. Can't,,, my projects are usually HUGE and require "Best Quality" setting which does not 'pre-render' unless I am not understanding what you mean. ..
QT/Full Quality, yepp, that's what I meant.. though, iDVD allows only a max. of 120min, the file couldn't be larger than 26GB... usual advice for video makers: more harddrive space, much more hdd space, moooooore...
as said. if you hand-over a complex iM project with lots of different files and formats, you increase the encoding time; and on my German board, I read the advice to avoid mp3 in iM projects, cause those could slow down iDVD dramatically... (one guy converts every audio import into aiff... before usage...) ...
but....
PC-guys LOVE to discuss 'speed', I never ever had any concerns about speed on my Mac..
as long as you don't render a CGI hairy KingKong or calculate the weather in the year 2020, speed is - imho - no criteria... as you said: a Mac product looks better, more professional, 'feels' better...
those 'nerds', watching their activity meters, are faster, so what - but what do they CREATE faster...? benchmarks... ??
give'em a break with those:
http://movies.apple.com/movies/us/apple/getamacads2/work480x376.mov
http://movies.apple.com/movies/us/apple/getamac/betterresults_480x376.mov
aaand 'Mac'/Justin Long is partner of Bruce Willis in http://www.imdb.com/title/tt0337978/ , Yippee Ki Yay M.. ;))

Speeding up rendering.

Similar Messages

Maybe you are looking for