Bug - Project Settings - Custom Mask area

[Intrigued]

This is a bug and suggestion post...

The Project Settings dialog area should (recommendation part) popup an error when no mask is set for when the Custom Mask option is chosen, instead of when the Preview/Build dialog comes up.

And...

When I did not choose a mask for the Custom Mask option and then clicked on the OK button... then tried to bring such dialog back up the app. took such a long time that the Not Responding windows error came up (see: attachment).

I have noticed this a couple times... AMS 6 seems to stall. I will try to remember where and when this happens to present such for the DEV folks.

[Intrigued]

Plus I have my Layout set so that the AMS 6 IDE minimizes when I Preview (F5) projects... .now... when I try to restore (maximize) the IDE after my previous post's actions... the IDE takes quite a bit (several to a couple dozen seconds to restore (maximize).

:huh

[Brett]

I have submitted the part about validating the mask filename. REF: 11064

However, I can't simulate either of the other "delay" issues at all. Can anyone else?

[yosik]

Difficult to say what is a delay and what isn't...
I don't feel a special delay here, but it may be related to disk speed/defrag/number of apps open/Antivirus program etc....Lots of factors...

Yossi

[Corey]

I've never defragmented my hard drive and it works just as fast today as the day I bought it a couple years ago... So I doubt that's a source of significant delay FWIW. :yes

[yosik]

Depends on how many times you delete files and save other files. In my experience, you can get almost a 50% performance boost. But if you don't do the delte/save cycle a lot, then defrag won't really help you.
Yossi

[Corey]

On my system I have been abusing this hard drive like crazy for over two years without a single defrag, including daily multitrack audio recording (and a ton of other hard drive intensive activity) and it is as speedy as the day I bought it, which is to say superb. Just a stock 7,200 RPM IDE drive too. I probably create/delete more files on a yearly basis than about 90% of computer users. Plus I've done a few huge batch jobs such as the button collection with over 200,000 files, etc. :)

This is identical to my experience with my second computer as well as the computer I used prior to this one, so I think the "defragmentation performance boost" is a bit of an urban myth as long as you buy good gear, but of course I'm certainly not questioning your experience Yossi, I'm sure it's different for every setup. In all 3 cases I use/used Maxtor Diamond 7,200 RPM IDE hard drives for years without a single hard drive related problem or noticeable change in speed. :yes

Now the VIA chipset in the old box, *that's* a whoooole different story, but I digress... Let's just say I don't miss the blue screen. :)

[Derek]

I wouldn't mind betting, Corey, that your HDD is slower than when you bought it 2 years ago and you haven't noticed it slowing down because of the time period.

I'd be tempted to say that defragging with a good defrag tool will show a difference. :yes

[Derek]

... and finally, i reach 1,000 posts [again!] :yes

[Corey]

Congrats on the millennium Derek! :yes
I wouldn't mind betting, Corey, that your HDD is slower than when you bought it 2 years ago and you haven't noticed it slowing down because of the time period.
It's possible but I doubt it, I use certain large VSTis on a daily basis which use DFD (Direct From Disk, see below) right at the threshold of system latency (<10ms). Any difference in hard drive speed would probably be immediately evident. If you check out the latest AMS Flash tour for example, that was played just a couple weeks ago using a 24bit sampled Bosendorfer piano at <10ms latency without a single glitch directly from a stock IDE hard drive which has been heavily abused for years and never defragmented. If there is a slight performance difference it's not affecting the speed of any critical processes so far... :yes

As for defragging, I don't think I can realize much of a boost given performance is pretty much identical to when I bought it. Bear in mind that it was a good quality computer, I had it custom built and the guy used all high quality parts/RAM. Also, I always keep my hard drive at least 25% empty, ideally 50%-60%, by backing up daily to an offline external drive. The only maintenance I've done was a recent replacement of the Northbridge and CPU fans. Anyhow I'm not challenging anyone's beliefs but my experience is that a good drive seems to do OK without defragmenting, or at least my opinion is that it's likely not the source of Yossi's reported delay anyhow... :yes


The separately available Direct from Disk extension enables KONTAKT to play samples directly from the hard drive. Sample size will no longer be limited by the amount of physical RAM - an instrument can be as large as available hard drive space. All KONTAKT instruments will load many times faster than RAM-based instruments.
http://news.harmony-central.com/Newp/2003/Kontakt-12.html

[TJ_Tigger]

... and finally, i reach 1,000 posts [again!] :yes
Congrats Derek . . . again . . .

[Lorne]

Fragmentation has nothing do with the quality of the drive, but rather the quality of the file system, and your usage patterns. (To some degree it is also affected by how full the partition is.)

NTFS in particular is remarkably resilient to fragmentation.

[Corey]

I've got the rootingest, tootingest usage habits this side of the Pecos. Just ask my life coach. One thing about daily digital multitrack recording is that it creates *tons* of fairly large files which are constantly just getting deleted, moved, or replaced anyhow. Absolutely tons. :yes

[Lorne]

Yeah. Deleting large files doesn't generally cause a lot of fragmentation, though. :)

BTW, here's a tip I picked up a while back for evaluating the impact of fragmentation on a partition:

Easy way to measure: In Performance Monitor, look at the "logical disk" counters for the partition of interest. Pay particular attention to the total IO rate -- "disk transfers/sec" -- and to "split IO/sec". Split IOs are "extra" IO requests resulting from IO requests that had to be broken up into pieces - either because the buffers were too large to handle in one transfer (very rare) or due to fragmentation of the files.

If your split IO rate isn't a significant fraction of the total transfers/sec, you don't have a problem that defragging will fix.

Let's say you are seeing an average of 100 IOs/sec but only 5 split IOs/sec - ok, defragging would get rid of about 5% of the IOs the disk is doing. As to whether that's "significant", that's your call. :) IMO, more than 30% split IOs would certainly be significant, less than 1% certainly not; in between, it depends on circumstances.

(IME an awful lot of time and disk bandwidth is wasted defragging files that are in fact very badly fragmented, but which are not actually accessed very often...)

[Corey]

Creating, deleting, and moving large numbers of files doesn't cause disk fragmentation? Cool, I never knew that... :yes

[Lorne]

Creating, deleting, and moving large numbers of files doesn't cause disk fragmentation? Cool, I never knew that... :yes
Large numbers of files == yes
Predominately large files == no

(And once again, in NTFS, it has to get really, really bad for it to matter.)

[Lorne]

I don't see any "logical disk" counters in Performance Monitor...Note: Performance Monitor, not Windows Task Manager. :)

Start > Programs > Administrative Tools > Performance
Right click on graph > Add Counters
Select the "LogicalDisk" performance object

[Corey]

Digital multitracking creates tons of files, most of which ended up getting deleted or moved. I do that daily. Anyhow, yeah I was in Performance Monitor, it won't let me add counters. Not sure what's up with that but on the other hand it's not really important so I plan to attack the problem slowly and lethargically over the course of the next decade and see if that turns up any solutions. :)

[csd214]

I've never defragmented my hard drive and it works just as fast today as the day I bought it a couple years ago...
I don’t question your feeling, but it’s hard to believe. Diskeeper Corp. has published some scientifically (?) documented tests about the consequences of fragmentation. It seems like Corey and Diskeeper Corp are the extreme points; personally I’m somewhere in the middle (my feeling only).

Later on I read, “I always keep my hard drive at least 25% empty, ideally 50%-60%, by backing up daily to an offline external drive.” I think that explains a lot.

Are you using System Restore? Do you run a desktop search tool (Google or MS)?

I think Yossi hit the nail with “Difficult to say what is a delay and what isn't... Lots of factors...”


As to the subject:
I can’t notify any delay as described by Intrigued. BUT: If I hit preview I get the build error dialog. I OBSERVE A DELAY WHEN I CLICK ‘CLOSE’ OR AT THE NEXT STEP; RESELECTING THE PROJECT MENU.

[Corey]

I'm not making any claims about other systems, and I don't know about 1-2% difference, but when I stretch my system to the limit using DFD (or any other software) it seems to work at exactly the same speed as the week I bought it. Also when I move files, they seem to go at exactly the same speed as they always have, and I move around *tons* of large folders all the time. Definitely nowhere near a double digit difference, I can guarantee it. And we're talking over a period of over 2 years of heavy daily usage, so... :)

One thing about using large, polyphonic VSTis with DFD in a low latency environment, you notice any significant difference in speed right away. You hear it, and you feel it. There no way around it. Sample libraries are getting *huge*. A double digit change in the speed of your drive would be very obvious at <10ms latency with any large VSTi using DFD, even over time. My personal sense of time, and ability to perceive changes in time, is fairly precise from decades spent specifically developing it through musical study. And I'm not judging this relatively, I am judging it absolutely against the reference point of my system under those conditions. I've played certain titles on several different systems, including Macs, so I'm reasonably familiar with the differences too. NI Elektrik Piano for example can buckle the knees of practically any system under the right circumstances... :yes

Anyhow this drive will be retired very soon and I doubt I'll be defragmenting the replacement, although I will take the time to benchmark it with a good old analog stopwatch so I can test my theory a couple years down the line. My interest is in the actual measurable difference during real world tasks, not the theoretical threshold of change. And I'd be willing to bet that I can use the new drive heavily for two years and still move big files around, etc. without any significant measurable loss of speed, i.e. <5%. We'll see, only time will tell. :yes

[yosik]

... and finally, i reach 1,000 posts [again!] :yes
Now that you said it, I noticed that I, too, have passed the millenium milestone.
:yes
Yossi

[Derek]

... and finally, i reach 1,000 posts [again!] :yes
LOL ... ohh this is a joke, man! Back to under 1,000 AGAIN.
Think i'm gonna register a brand new name and try again. :rolleyes

[yosik]

Hey!
What's going on?? I am also back to 800 plus posts....
Is there a robot which lowers the post numbers??????

Yossi

[TJ_Tigger]

I believe with the AMS6 beta forum being gone, those posts no longer exist and therefore do not count in your total posts anymore. I had made it over the 2k mark with the beta forum and now am back in the 1900s. Oh well. We all could continue to post here and inflate our posts. :D

Tigg