Welcome to the Piano World Piano Forums
Over 2 million posts about pianos, digital pianos, and all types of keyboard instruments
Join the World's Largest Community of Piano Lovers (it's free)
It's Fun to Play the Piano ... Please Pass It On!

SEARCH
the Forums & Piano World

This custom search works much better than the built in one and allows searching older posts.
(ad) Pianoteq
Latest Pianoteq add-on instrument: U4 upright piano
(ad) Pearl River
Pearl River Pianos
(ad) P B Guide
Acoustic & Digital Piano Guide
PianoSupplies.com (150)
Piano Accessories Music Related Gifts Piano Tuning Equipment Piano Moving Equipment
We now offer Gift Certificates in our online store!
(ad) Estonia Piano
Estonia Piano
Quick Links to Useful Stuff
Our Classified Ads
Find Piano Professionals-

*Piano Dealers - Piano Stores
*Piano Tuners
*Piano Teachers
*Piano Movers
*Piano Restorations
*Piano Manufacturers
*Organs

Quick Links:
*Advertise On Piano World
*Free Piano Newsletter
*Online Piano Recitals
*Piano Recitals Index
*Piano Accessories
* Buying a Piano
*Buying A Acoustic Piano
*Buying a Digital Piano
*Pianos for Sale
*Sell Your Piano
*How Old is My Piano?
*Piano Books
*Piano Art, Pictures, & Posters
*Directory/Site Map
*Contest
*Links
*Virtual Piano
*Music Word Search
*Piano Screen Saver
*Piano Videos
*Virtual Piano Chords
Page 2 of 2 < 1 2
Topic Options
#1996511 - 12/08/12 05:16 PM Re: Galaxy Vintage D - Audio Dropouts ??? [Re: Macy]
Acca Offline
Full Member

Registered: 11/10/12
Posts: 67
Originally Posted By: Macy
I play Kontakt for hours with 8 GB of RAM and it never pages-out and there are no drop-outs.


I can play Vintage D for hours on a 2Gb Ram, 6 year old laptop with an old dual core processor, and it will not drop out. Granted it's at the limit of RAM space and when resuming from sleep that sometimes screws it up.

If the OP thinks that changing to a firewire/USB 3 drive will help, he will be disappointed.

Top
(ad) Roland

Click Here

#1996517 - 12/08/12 05:33 PM Re: Galaxy Vintage D - Audio Dropouts ??? [Re: musicmad]
sullivang Offline
2000 Post Club Member

Registered: 07/05/09
Posts: 2100
Loc: Sydney, Australia
Macy:
Ok. I just know that on Windows, it definitely relies on the file system cache for caching that portion of the sample that occurs after the pre-load portion. I know this because when I clear the file system cache, and hold a note for a long time, it re-loads that later portion of the sample from disk. If I don't clear the file system cache, playing the same note twice, and holding it for a long time, does NOT result in any disk activity the second time it is played.

So, given that it is only using the file system cache for the later portion of the samples, it seemed logical to me that this memory would be freed up if an application really did attempt to allocate some memory proper. (but as I later edited, it's still possible that this doesn't happen immediately)

I realise that the Mac design could be totally different, in any case.

Yes, agreed, if there is paging that occurs when the dropouts occur, of course that would be a likely cause.

Greg.


Edited by sullivang (12/08/12 05:47 PM)

Top
#1996527 - 12/08/12 05:50 PM Re: Galaxy Vintage D - Audio Dropouts ??? [Re: sullivang]
Macy Offline
500 Post Club Member

Registered: 03/09/10
Posts: 563
Originally Posted By: sullivang
So, given that it is only using the file system cache for the later portion of the samples, it seemed logical to me that this memory would be freed up if an application really did attempt to allocate some memory proper. (but as I later edited, it's still possible that this doesn't happen immediately)

Yep, that is what I said. The Inactive RAM does not get paged-out until more active RAM is needed. Otherwise the inactive RAM simply remains filled with the previous data. So if the application needs that data again, it will become active RAM again, without reading the disc.
_________________________
Macy

CVP-409GP, Vintage D, Ivory II GP's & American Concert D, Ravenscroft 275, True Keys American D, Garritan Authorized Steinway, Alicia's Keys, EWQL Pianos, MainStage, iPad/forScore/PageFlip Cicada, Custom Mac MIDI/Audio Software Design, Macs Everywhere

Top
#1996538 - 12/08/12 06:22 PM Re: Galaxy Vintage D - Audio Dropouts ??? [Re: musicmad]
sullivang Offline
2000 Post Club Member

Registered: 07/05/09
Posts: 2100
Loc: Sydney, Australia
I'm confused now. To me, "Inactive RAM" is memory for which the data is no longer required. That memory is now available for anything - the same application, or another application. I thought that "Inactive RAM" would NEVER be paged out, because the data in that memory is not needed any more. I thought only "active" RAM would ever be paged out - paging saves the contents of memory to disk, so that that same block of RAM can be used for something else. When the paged out memory is required again, it is again loaded into RAM somewhere.

If you're saying that there is a distinction between "Inactive RAM" and memory that has been deallocated completely, then that's a subtlety I was not aware occurred.

Greg.


Edited by sullivang (12/08/12 06:32 PM)

Top
#1996548 - 12/08/12 06:46 PM Re: Galaxy Vintage D - Audio Dropouts ??? [Re: musicmad]
MacMacMac Offline
3000 Post Club Member

Registered: 09/24/09
Posts: 3672
Loc: North Carolina
sulli: You terminology is not quite right (active/inactive), but you have the concept down correctly ... at least for the heap space: Unused memory will not be paged out. What would be the point? After all, heap space is just memory used by a process (or shared by multiple processes). If the process(es) have exited, the content of the memory it/they formerly owned is pure junk which cannot be used for any purpose. So there's no need to page it out. Instead, the memory can be reallocated as needed.

But in the disk buffer cache things are a bit different. Any disk reads or writes coming into the buffer cache will stay there, possibly for a long time. Even if the process(es) that fetched from (or wrote to) a file that is resident in the buffer cache have terminated, the cache content remains valid. So, a new process can open a file whose contents are already resident in the buffer cache ... and the cached content can be used instead of fetching from disk.

I don't think any of the buffer cache ever gets paged out. What would be the point? If space is needed in the buffer cache for new disk reads/write and none is available, some of the existing content can either be (a) discarded if it is not "dirty", or (b) copied to the disk file(s) to which it belongs, if it is "dirty". Either way, this frees up some buffer cache space. No need to swap/page it at all.

This is my understanding of Unix, and so it should apply to the Mac. It might be partly or substantially true for Windows, too ... not sure. But the OP has a Mac, so ...

Top
#1996558 - 12/08/12 07:13 PM Re: Galaxy Vintage D - Audio Dropouts ??? [Re: sullivang]
Macy Offline
500 Post Club Member

Registered: 03/09/10
Posts: 563
Originally Posted By: sullivang
I'm confused now. To me, "Inactive RAM" is memory for which the data is no longer required. That memory is now available for anything - the same application, or another application. I thought that "Inactive RAM" would NEVER be paged out, because the data in that memory is not needed any more. I thought only "active" RAM would ever be paged out - paging saves the contents of memory to disk, so that that same block of RAM can be used for something else. When the paged out memory is required again, it is again loaded into RAM somewhere.

If you're saying that there is a distinction between "Inactive RAM" and memory that has been deallocated completely, then that's a subtlety I was not aware occurred.

Greg.


Like I said I didn't really want to go into this because the terminology is confusing, the Mac is not like a PC, and even though the Mac uses Unix the terminology is a little different between them too.

On a Mac:

Wired RAM - is memory that is used by the OS and never paged-out and never made inactive.

Active RAM - is memory that is currently in use by a currently active application. It is never paged-out, but it can be made inactive RAM.

Free RAM - RAM that has not been used yet.

Inactive RAM - is memory that is NOT currently in use but was in use by a currently active application or the OS. It is data kept in RAM so that if a current application wants the data again it is already in RAM and doesn't need to be read from the disc again. BUT if the current application, or the system, needs more Active RAM that exceeds the available Free RAM (the amount of RAM the OS determines must remain free) then Inactive RAM is paged-out to the disc.

Data is not "deallocated completely" from inactive RAM until its application quits or the RAM was paged-out to make additional active RAM. That way any current app can get the data again that it has previously read as long as it wasn't paged-out.

OK?
_________________________
Macy

CVP-409GP, Vintage D, Ivory II GP's & American Concert D, Ravenscroft 275, True Keys American D, Garritan Authorized Steinway, Alicia's Keys, EWQL Pianos, MainStage, iPad/forScore/PageFlip Cicada, Custom Mac MIDI/Audio Software Design, Macs Everywhere

Top
#1996561 - 12/08/12 07:17 PM Re: Galaxy Vintage D - Audio Dropouts ??? [Re: gvfarns]
o0Ampy0o Offline
Full Member

Registered: 11/18/12
Posts: 458
Originally Posted By: gvfarns
Brain damage. You certainly implied exactly what I said you did. If you had previously inferred it, that's also wrong, but it's your own personal problem. Had you not shared that incorrect inference here, no one else would have been subjected to it and there would be no confusion on the matter.

But anyway, if you did not intend to say what I quoted you as saying, then that's good. It's just a failure to communicate. If anyone wants to know the answer to the question musicmad posed, they can ignore your posts and get the correct info from the other posters in this thread.


This is being addressed in a separate thread here.

Top
#1996687 - 12/09/12 01:10 AM Re: Galaxy Vintage D - Audio Dropouts ??? [Re: musicmad]
sullivang Offline
2000 Post Club Member

Registered: 07/05/09
Posts: 2100
Loc: Sydney, Australia
Macy: Thankyou! Much much better. I have done a bit of Googling as well, and it is all becoming clear. All those terms are new to me.

I was thinking that the application had taken some action to make Active RAM become Inactive RAM, such as deallocating virtual memory. The only "action" that is taken is INACTION. The OS deems those pages of memory as good candidates for swapping out simply because the application has not used those pages for a while! smile

A couple of points though:
1. I assume Free Memory could also be memory that was in use, but has now become free, due for example to an application explicitly deallocating virtual memory, or to the application closing. (agreed?)

2. This statement of yours does not seem to be quite right:
Originally Posted By: Macy

Originally Posted By: Sullivang
So, given that it is only using the file system cache for the later portion of the samples, it seemed logical to me that this memory would be freed up if an application really did attempt to allocate some memory proper. (but as I later edited, it's still possible that this doesn't happen immediately)


Yep, that is what I said. The Inactive RAM does not get paged-out until more active RAM is needed. Otherwise the inactive RAM simply remains filled with the previous data. So if the application needs that data again, it will become active RAM again, without reading the disc.


IMHO it is invalid to refer to the file system cache as "Inactive RAM", because pages that the OS has used to cache disk reads would NEVER be paged out - they would simply be removed from the cache and allowed to be allocated. (presumably entering the Active state at the outset if they are to be used immediately, or being marked as Free) I.e - it doesn't make sense to write the data out to the page file, because the same data is already in the sample file from which it was read. ;^)

I know we haven't yet determined whether the file system cache is used for Kontakt on a Mac.

Greg.


Edited by sullivang (12/09/12 01:56 AM)

Top
#1996762 - 12/09/12 06:39 AM Re: Galaxy Vintage D - Audio Dropouts ??? [Re: musicmad]
MacMacMac Offline
3000 Post Club Member

Registered: 09/24/09
Posts: 3672
Loc: North Carolina
In the Unixes, the file system cache is used for all file accesses. (To bypass it would require a system hack.)

Top
#1996763 - 12/09/12 06:43 AM Re: Galaxy Vintage D - Audio Dropouts ??? [Re: musicmad]
sullivang Offline
2000 Post Club Member

Registered: 07/05/09
Posts: 2100
Loc: Sydney, Australia
MacMacMac: I do not think that is correct. It is possible to do direct i/o to files which bypasses the cache. (and I don't mean raw i/o to the disk - it is still file level i/o)

Greg.

Top
#1996767 - 12/09/12 06:55 AM Re: Galaxy Vintage D - Audio Dropouts ??? [Re: musicmad]
MacMacMac Offline
3000 Post Club Member

Registered: 09/24/09
Posts: 3672
Loc: North Carolina
Well, that must be a special case. I only know of raw-disk access. Used it for a while back in the 80s for performance on database servers.

But whether by raw-disk or other means, does this piano software bypass the cache? And if so, why would it? You play the same notes on the piano again and again, so you need the same data repeatedly. The disk cache is a fine way to accomplish that.

If the software bypasses the cache, it would have to implement its own caching. Why invent what's already there?

Top
#1996771 - 12/09/12 07:01 AM Re: Galaxy Vintage D - Audio Dropouts ??? [Re: musicmad]
sullivang Offline
2000 Post Club Member

Registered: 07/05/09
Posts: 2100
Loc: Sydney, Australia
MacMacMac: As I said, on Windows the samples are read through the file system cache - I have proved it IMHO. Chances are it's the same situation on the Mac, but we don't know for sure yet. If it DOES, then to re-iterate my original point: it seems plausible that if Kontakt has caused the file cache to grow appreciably, and then something (another app, or even Kontakt itself) needs to allocate some memory, rather than page out memory that is already in use, it would probably use some memory from the file cache instead, which would cause no disk activity at all. (until of course the sample data that was in those pages of memory is needed again)

I agree that it is more likely than not that the file cache is used for Kontakt on the Mac.

Greg.

Top
#1996773 - 12/09/12 07:12 AM Re: Galaxy Vintage D - Audio Dropouts ??? [Re: musicmad]
sullivang Offline
2000 Post Club Member

Registered: 07/05/09
Posts: 2100
Loc: Sydney, Australia
(and to prove that files can be accessed without the cache on Unixes, you could try the fio file system exerciser. See the "direct=[true|false]" flag. I have tried this on Ubuntu. )

Greg.


Edited by sullivang (12/09/12 07:17 AM)

Top
#2070015 - 04/24/13 02:50 PM Re: Galaxy Vintage D - Audio Dropouts ??? [Re: musicmad]
sorrownightingale Offline
Junior Member

Registered: 07/10/12
Posts: 6
Loc: Tenerife-Canary Island-Spain
Hi,
I am running Vintage D 1.2 on Kontakt 5 full, my system is an Intel Quad Core Q9400 2.66Ghz 8GB ram winx64 with Creative Audigy 4 pro, asio 24/96, latency 7.0 mseg, buffer 672 samples, pre-load 240KB. My keyboard Yamaha S90xs and CME GPP-3 tri pedal.On Kontakt 4 I did not have drop out but, with Kontakt 5 if I enable repedalling on Vintage D, cpu load increases and then drop out, maybe,it is too much information midi of the continuos sustain pedal.I have run latency check and my system is enabled, then, what is the problem?

Top
Page 2 of 2 < 1 2

Moderator:  Piano World 
What's Hot!!
HOW TO POST PICTURES on the Piano Forums
-------------------
Sharing is Caring!
About the Buttons
-------------------
Forums Rules & Help
-------------------
ADVERTISE
on Piano World

The world's most popular piano web site.
-------------------
PIANO BOOKS
Interesting books about the piano, pianists, piano history, biographies, memoirs and more!
(125ad) Dampp Chaser
Dampp Chaser Piano Life Saver
Sheet Music
(PW is an affiliate)
Sheet Music Plus Featured Sale
Download & Print Sheet Music Instantly
sheet music search
sheet music search

sheet music search
(ad) HAILUN Pianos
Hailun Pianos - Click for More
(ad) Lindeblad Piano
Lindeblad Piano Restoration
Who's Online
146 registered (ADWyatt, Almaviva, 36251, AmateurBob, 42 invisible), 1737 Guests and 38 Spiders online.
Key: Admin, Global Mod, Mod
Forum Stats
74270 Members
42 Forums
153629 Topics
2251683 Posts

Max Online: 15252 @ 03/21/10 11:39 PM
New Topics - Multiple Forums
CA95 noisy front rail
by lolatu
28 minutes 56 seconds ago
It's not so simple as the 10,000 hour rule.
by Brad Hoehne
42 minutes 37 seconds ago
Day brightener
by Chuck Behm
Today at 10:09 AM
Paris Atelier
by Beemer
Today at 09:20 AM
Lubricious Whippen Cushions
by Grotriman
Today at 09:12 AM
(ads by Google)

Visit our online store for gifts for music lovers

 
Our Piano Related Classified Ads
| Dealers | Tuners | Lessons | Movers | Restorations | Pianos For Sale | Sell Your Piano |

Advertise on Piano World
| Subscribe | Piano World | PianoSupplies.com | Advertise on Piano World | Donate | Link to Us | Classifieds |
| |Contact | Privacy | Legal | About Us | Site Map | Free Newsletter | Press Room |


copyright 1997 - 2014 Piano World ® all rights reserved
No part of this site may be reproduced without prior written permission