136 postsPage 3 of 7
1, 2, 3, 4, 5, 6, 7
admiralbumblebee wroteYou don't appear to understand what the issue is. You're discussing latency, the video is not discussing latency.

I feel like niles understood and addressed this.
niles wroteIn 3.5.1 the safety buffer was removed which lowered the latency by roughly 25%, trade off is jitter due to buffer changes when monitoring.

Mac OS X Catalina 10.15.7
Mac Pro 6.1
3 GHz 8-Core Intel Xeon E5
32 GB 1066 MHz DDR3
Dual AMD FirePro D500 3072 MB
Quantum 2
User avatar
by Jemusic on Sun Jul 12, 2020 5:54 pm
Firstly this is not about how accurately Studio One records and plays back midi timing pointing to external instruments. I have shown earlier that it does that very well. Down to fractions of a millisecond and it plays back that way especially over a midi system that is PCI based. Something most people wont be doing.

But this test is showing up timing inaccuracies in terms of how well a virtual instrument plays back. Firstly I thought the test was unfair because the Admiral was using Sample One inside Studio One and something else inside Ableton.

I tried a range of different virtual instruments inside Studio One and got a very similar result to the Admiral here. I used a digital metronome feeding the AKAI ME35T which converts audio to midi notes and very accurately might I add. Timing was close but did vary (worst case) as much as 15 mS between the audio hits that Sample One generated. I used a very fast and tight click type sound as well as the Sample One sound. In the other virtual instruments I used fast rise time percussive type sounds and got a very similar result so it is not based on the virtual instrument itself.

I ran the same test in Ableton. Ableton is tighter, more consistent and better in this regard. By the way Ableton's roundtrip latency depends on what you are using hardware wise. Mine is 3 mS roundtrip with a Focusrite Clarett interface over thunderbolt. Studio One though is just over 1mS for me with the same interface.

Audio can be quantised and made totally perfect in Studio One but the situation is different for those who may be playing very accurately and subtly against the click and want it played back exactly the same way.

Specs i5-2500K 3.5 Ghz-8 Gb RAM-Win 7 64 bit - ATI Radeon HD6900 Series - RME HDSP9632 - Midex 8 Midi interface - Faderport 2/8 - Atom Pad/Atom SQ - HP Laptop Win 10 - Studio 24c interface -iMac 2.5Ghz Core i5 - High Sierra 10.13.6 - Focusrite Clarett 2 Pre & Scarlett 18i20. Studio One V5.5 (Mac and V6.5 Win 10 laptop), Notion 6.8, Ableton Live 11 Suite, LaunchPad Pro
User avatar
by admiralbumblebee on Sun Jul 12, 2020 5:58 pm
robertgray3 wrote
admiralbumblebee wroteYou don't appear to understand what the issue is. You're discussing latency, the video is not discussing latency.

I feel like niles understood and addressed this.
niles wroteIn 3.5.1 the safety buffer was removed which lowered the latency by roughly 25%, trade off is jitter due to buffer changes when monitoring.


He is describing and showing latency, not jitter. In all of my tests (as I very clearly describe in the text article!) the latency is compensated for manually. There is a partial list of these considerations here: http://admiralbumblebee.com/music/2020/ ... itigations

He also brings up a change in Studio One v3, despite this issue very clearly existing in Studio One 2 as per prior documentation and discussion.

As I mentioned before, you can directly look at the events streaming through S1 by using a plugin to monitor them and see that the offset data is not being sent. This has nothing to do with safety buffers.

Even if we accepted the (erroneous) assumption that this is a safety buffer issue, there are 3 other DAWs that pass this test perfectly and also use a safety buffer system. I will be covering these in the next video.

It's entirely possible that there is a complicating factor regarding Studio One's buffer system internally. The existence of the issue before the introduction of it clearly demonstrates that it's not the cause of it however.

I hope I'm not coming off as rude or combative. I'm attempting to be direct and clear regarding this information to minimize misunderstandings.

Please note as well that I run NO ADS on my website or videos. I post my finances publicly regarding the website/videos and I lose a few hundred to a few thousand a year doing this. (Not to mention thousands of hours!)

User avatar
by Rasi on Sun Jul 12, 2020 6:15 pm
Jemusic wroteAudio can be quantised and made totally perfect in Studio One but the situation is different for those who may be playing very accurately and subtly against the click and want it played back exactly the same way.


I’m sorry, but this seems to be consistently brought up by you specifically over the years after searching for answers myself on this forum about how to remedy S1’s issue with midi. “Utilize input quantize”, or “maybe disable input quantize”... Now it’s “quantize your audio” ...

Are we not recording music? When did it become a requirement that in order to multi-track and record *music* in the modern age, using a DAW like S1, you should expect to erase all nuance and aspects of performance. “For those who may be playing very accurately.” What kind of subtle dig is this? I want to hear what I played!. Period.

How about: If YOU want to erase all nuance and performance, and YOU want to use quantization for all recorded midi and audio... then you can choose to do as such. For the rest of us, we track it in, and if we don’t get it right then we try again. Even better we can choose to comp multiple takes (which studio one does very well :thumbup: ) and build ourselves a better performance. All without resorting to a crutch.
User avatar
by Jemusic on Sun Jul 12, 2020 6:27 pm
@Rasi you missed my point completely. I am saying exactly that. Yes for those who want hard quantised parts this can easily be fixed but for those who are not doing that and want all the subtle nuances of the input timing, this needs to be improved for sure.

Specs i5-2500K 3.5 Ghz-8 Gb RAM-Win 7 64 bit - ATI Radeon HD6900 Series - RME HDSP9632 - Midex 8 Midi interface - Faderport 2/8 - Atom Pad/Atom SQ - HP Laptop Win 10 - Studio 24c interface -iMac 2.5Ghz Core i5 - High Sierra 10.13.6 - Focusrite Clarett 2 Pre & Scarlett 18i20. Studio One V5.5 (Mac and V6.5 Win 10 laptop), Notion 6.8, Ableton Live 11 Suite, LaunchPad Pro
User avatar
by Rasi on Sun Jul 12, 2020 6:42 pm
Jemusic wrote@Rasi you missed my point completely. I am saying exactly that. Yes for those who want hard quantised parts this can easily be fixed but for those who are not doing that and want all the subtle nuances of the input timing, this needs to be improved for sure.


My apologies for any misunderstanding. Indeed, I’ll be paying for v5 once there’s some proof of a fix because while Live has a few things going for it at the moment... I have many years experience with S1. Not true with any other DAW
User avatar
by roland1 on Sun Jul 12, 2020 7:58 pm
I will keep repeating myself so that we do not lose track of the issue. It has to do with MIDI input from an external source (keyboard, V-Drums, etc.) and how accurately Studio One (from v1-v5) is able to record that information and place it within the bar EXACTLY HOW IT WAS PLAYED.

F••• quantizing my natural, human grooves and sense of timing. I do that as a touch up only, not as a writing technique.

I have a free license for Ableton Live Lite via Focusrite. I'll try that out and see if it tracks my drumming more accurately or does away with the +/- 3 ticks when recording MIDI output from an instrument track.

Studio One Pro (v5) on i7 7700 win10 PC w16GB RAM and a Mac Pro Tower (w/RME & Focusrite interfaces.)

I use S1 as an author/musician/multi-media artist.
My work includes the newly released: Clearing a Path to Joy (And finding contentment along the way) [AuroraSkyPublishing.com]
and my upcoming music video, Too Big To Fail, which introduces Citizen Based Social Planning — "the next step in the evolution of democracy." You know...typical everyday stuff. :)
User avatar
by Jemusic on Sun Jul 12, 2020 8:15 pm
roland1 wroteI will keep repeating myself so that we do not lose track of the issue. It has to do with MIDI input from an external source (keyboard, V-Drums, etc.) and how accurately Studio One (from v1-v5) is able to record that information and place it within the bar EXACTLY HOW IT WAS PLAYED.


No, its not about this at all. I have already shown that Studio One can very accurately record incoming midi data and play it back. (especially to external instruments from a midi interface connected over a PCI port)

The real issue is how accurately a virtual instrument consistently plays back the recorded midi data and the resulting sound the virtual instrument accurately reflects that midi data. What we are seeing now is timing inconsistencies from the virtual instrument. eg If a series of very accurately placed midi notes eg every crotchet (either from an input such as the Admiral's video or a midi track that has quantised notes on it) the resultant virtual instrument output will fluctuate timing wise up to 15 mS differences or so between every crotchet. Some on the money, some early, some late etc..

But yes for your purposes its the end result that might make a drum part sound a little sloppy compared to the way you played it in. They are two different things. What we want now is the same response from a virtual instrument being triggered via midi. If a sound takes say 2mS to be generated from a midi note on, it needs to be 2mS for every midi note on.

Specs i5-2500K 3.5 Ghz-8 Gb RAM-Win 7 64 bit - ATI Radeon HD6900 Series - RME HDSP9632 - Midex 8 Midi interface - Faderport 2/8 - Atom Pad/Atom SQ - HP Laptop Win 10 - Studio 24c interface -iMac 2.5Ghz Core i5 - High Sierra 10.13.6 - Focusrite Clarett 2 Pre & Scarlett 18i20. Studio One V5.5 (Mac and V6.5 Win 10 laptop), Notion 6.8, Ableton Live 11 Suite, LaunchPad Pro
User avatar
by pawelhebda on Mon Jul 13, 2020 1:44 am
Here it is. 10 years waited MIDI sync accuracy in new Studio One 5. Ohh come on guys!!!

https://www.youtube.com/watch?v=ioxyBmR ... e=youtu.be
User avatar
by niles on Mon Jul 13, 2020 1:47 am
admiralbumblebee wroteYou don't appear to understand what the issue is. You're discussing latency, the video is not discussing latency.
I'm well aware of the inconsistencies due to buffer changes when monitoring virtual instruments in real time. If you process large chunks of data (high buffer setting) in real time without an extra buffer you get partially filled buffers when they are flushed which introduces the irregularities you are hearing and seeing. By adding an additional safety buffer that is fully filled before it's flushed will lead to a consistent data stream at the cost of extra latency.

You have a valid point, but it's not a bug it's a choice.

Here's an old but nice read about buffer and latency jitter.
https://www.archiv.rme-audio.de/old/eng ... _latec.htm

Below is an image of a test I did in 2016 where I lined up the results of a few DAW's showing their difference in consistency when monitoring at 128/44.1kHz. Take a close look at the relation between the Note-On message and the audio signal.

Image

As you can see the time between the Note-On and audio start in Reaper and Studio One are consistent and Cubase acts like Studio One is acting now. Back then Studio One latency was much higher than Cubase's latency, BUT the latency in Studio One was consistent. That changed in 3.5.1. If that's good or bad is highly subjective and depending on how you use Studio One (large or small buffer).

The PDF shows the exact figures behind wat you see in the image. The 'Audio latency compared
to MIDI Event' is important here, because it shows the consistency (and trade off).

Attachments
DAW_Latency_Test_128_44100_24112016.pdf
Figures behind the image
(91.35 KiB) Downloaded 114 times
Bitwig_Cubase_Reaper_DP_Studio_One_8_beat_RT_test_128_44100.gif

OS: Windows 11 Pro | HW: Gigabyte Z690-UD-DDR4 • INTEL i7 12700K • 64GB • 3x EVO 860 • NVIDIA GT1030 (@WQHD) • RME AIO
User avatar
by musicchamber on Mon Jul 13, 2020 4:20 am
pawelhebda wroteHere it is. 10 years waited MIDI sync accuracy in new Studio One 5. Ohh come on guys!!!

https://www.youtube.com/watch?v=ioxyBmR ... e=youtu.be


This is a perfect example of what I tested sometime back in the forum (but with using a metronome click and the same metronome click rendered. The timing is jittering all over the place.

I did the same test in Logic. Pro, no jitter / sync up problems, everything was in perfect sync.

There is a real problem here, and Presonus are ignoring this. It's not a new issue, it's been happening for ages!

Don't get me wrong, I love Studio One. Timing has got to be one of the top priorities in my opinion.

I also did another test syncing Studio One to an Atari ST running notator, and Studio One would not stay in sync with the click output from the Atari. I trust the Atari's and Notator timing over Studio One every time!

I can't see this problem going away personally, I think they are ignoring the issue. I don't know the in's and out's, I'm a musician not a technician, but I sure as hell know that I want perfect timing / sync with my DAW, and Studio One is like a jittery 80 year old with Parkinsons disease. I mean no offence to anyone who has Parkinsons but it is comparative.

I do not want to be argumentative, I'm only stating how I feel about this, and wish no one any harm in expressing myself.

Best,
Scott

Studio One Professional v6, Apple iMac Pro 14 core, 128gb memory, 4TB SSD. macOS Ventura, Main audio interface is RME Babyface Pro. Presonus 192, DP88 (not currently used), RC500, ADL600, Focusrite ISA430, TC4000 reverb, SPL Phonitor, Monitoring: Event Opal, IK Multimedia MTM Studio Monitors x 5, DMAX audio Super Cubes. HD800s mastering reference headphones. Sequential Prophet 12, Prophet 12 modules x4, E-MU 4XT Ultra, Roland Fantom 8, Korg Pa3x, Roland Fp7-F, MOTU midi express 128, Xkey air 37, Studiolive CS18, Atari STE with Notator plus loads of microphones and plugins etc
User avatar
by pawelhebda on Mon Jul 13, 2020 5:44 am
They have their own remote UCNET protocol!!! Why not implement it for syncing Studio One to another SO or NOTION??? Does they even tested it before releasing?
User avatar
by leosutherland on Mon Jul 13, 2020 6:25 am
musicchamber wroteI also did another test syncing Studio One to an Atari ST running notator, and Studio One would not stay in sync with the click output from the Atari. I trust the Atari's and Notator timing over Studio One every time!

Scott, you still running Notator on Atari?! :D

How it comparing with S1?

I used Notator on Atari 1040STE back in the day - it was a great tool then!

...said Halo

Studio One Pro v4.6.2 / v5.5.2 / v6.5.2
Serum, Diva, Repro, Synthmaster, Syntronik Bully, MTron Pro, Kontakt 6/7, AIR synths, Cherry Audio synths, Battery 4, PPG Wave 3.V, Generate

3XS SQ170 Music Studio PC
Windows 10 x64 (22H2)
i7 8700 Hexcore 3.2GHz, 16GB, 2TB 970 EVO+ M.2 NVMe SSD + 1TB SATA HD
Scarlett 2i4 (G2), Korg Taktile 25, Faderport 2018, ATOM


Beyerdynamic DT990 Pros, JBL 305P II speakers
User avatar
by roland1 on Mon Jul 13, 2020 7:13 am
Jemusic wroteNo, its not about this at all. I have already shown that Studio One can very accurately record incoming midi data and play it back.


I'm not a violent person, but I want to bludgeon you with a 50 lb bag of birdseed. READ my first post, which contains such concrete topic conveyances such as: "If you are concerned about accuracy of MIDI performance, I've put this test together to give us a common means of measuring the accuracy of MIDI input from system to system."

I have called you predictable more than once (including in my post when I said "I was expecting you...") in that each time I post something negative about Studio One, you're first to show up, like the press secretary standing on the White House lawn and trying to deflect blame for what's going on.

I want accuracy in MIDI notes WHEN they're being recorded. THIS HAS NOTHING TO DO WITH VIRTUAL INSTRUMENTS! WHY DON'T YOU WANT TO UNDERSTAND THAT? [Eyes welling up with tears, cry-screaming at high pitch.]

Admiral Bumblebee showed you CLEARLY in the video that there is a problem with INPUT recording called JITTER. That's what we're addressing in this thread. If your system works fine, that's beautiful. I'll give you my mailing address and you can send it to me here in Canada. In the meantime, I and many others must try to make our systems work in accord with whichever version of Studio One we own at present. I just bought v5, so the counter intel is wasted on a guy like me.

So please stop contradicting MY intentions for this post. ($&#!!@)

Studio One Pro (v5) on i7 7700 win10 PC w16GB RAM and a Mac Pro Tower (w/RME & Focusrite interfaces.)

I use S1 as an author/musician/multi-media artist.
My work includes the newly released: Clearing a Path to Joy (And finding contentment along the way) [AuroraSkyPublishing.com]
and my upcoming music video, Too Big To Fail, which introduces Citizen Based Social Planning — "the next step in the evolution of democracy." You know...typical everyday stuff. :)
User avatar
by admiralbumblebee on Mon Jul 13, 2020 7:20 am
niles wrote
admiralbumblebee wroteYou don't appear to understand what the issue is. You're discussing latency, the video is not discussing latency.
I'm well aware of the inconsistencies due to buffer changes when monitoring virtual instruments in real time. If you process large chunks of data (high buffer setting) in real time without an extra buffer you get partially filled buffers when they are flushed which introduces the irregularities you are hearing and seeing. By adding an additional safety buffer that is fully filled before it's flushed will lead to a consistent data stream at the cost of extra latency.

You have a valid point, but it's not a bug it's a choice.

Here's an old but nice read about buffer and latency jitter.
https://www.archiv.rme-audio.de/old/eng ... _latec.htm

Below is an image of a test I did in 2016 where I lined up the results of a few DAW's showing their difference in consistency when monitoring at 128/44.1kHz. Take a close look at the relation between the Note-On message and the audio signal.

Image

As you can see the time between the Note-On and audio start in Reaper and Studio One are consistent and Cubase acts like Studio One is acting now. Back then Studio One latency was much higher than Cubase's latency, BUT the latency in Studio One was consistent. That changed in 3.5.1. If that's good or bad is highly subjective and depending on how you use Studio One (large or small buffer).

The PDF shows the exact figures behind wat you see in the image. The 'Audio latency compared
to MIDI Event' is important here, because it shows the consistency (and trade off).


Ok, so it's clear I had a misunderstanding because we are getting different results. I understand the mechanism that you're explaining, but my results indicate that the change of the safety buffer mechanism did not change the jitter.

I was not getting a fixed latency in older S1 versions, nor was the other user that I worked with testing this, nor was the prior work on this that I found.

I'll take some time to go back into Windows and pull up some older versions and repeat the tests as such.

EDIT: I just now noticed when checking this on mobile that your image is animated. (I have this turned off on desktop). That makes your idea more clear. However, there is still the differing of our results, which I'll happily work on today.
Last edited by admiralbumblebee on Mon Jul 13, 2020 8:05 am, edited 1 time in total.

User avatar
by roland1 on Mon Jul 13, 2020 7:39 am
Thanks Scott, I'm glad to see that you "get it" in regards to this topic.

If this problem is going to get fixed (in the next ten years), we can't have others pretending that there is no problem. I expect such behavior in the world of politics, but not music recording software.

musicchamber wrote
This is a perfect example of what I tested sometime back in the forum (but with using a metronome click and the same metronome click rendered. The timing is jittering all over the place.

...

Best,
Scott

Studio One Pro (v5) on i7 7700 win10 PC w16GB RAM and a Mac Pro Tower (w/RME & Focusrite interfaces.)

I use S1 as an author/musician/multi-media artist.
My work includes the newly released: Clearing a Path to Joy (And finding contentment along the way) [AuroraSkyPublishing.com]
and my upcoming music video, Too Big To Fail, which introduces Citizen Based Social Planning — "the next step in the evolution of democracy." You know...typical everyday stuff. :)
User avatar
by musicchamber on Mon Jul 13, 2020 8:11 am
leosutherland wrote
musicchamber wroteI also did another test syncing Studio One to an Atari ST running notator, and Studio One would not stay in sync with the click output from the Atari. I trust the Atari's and Notator timing over Studio One every time!

Scott, you still running Notator on Atari?! :D

How it comparing with S1?

I used Notator on Atari 1040STE back in the day - it was a great tool then!


I have not had it set up for about 6 months, but I was sequencing some patterns with it linked up to my Korg and it is so great! I will have to set it up again because I do miss my little Atari :(

Best,
Scott

Studio One Professional v6, Apple iMac Pro 14 core, 128gb memory, 4TB SSD. macOS Ventura, Main audio interface is RME Babyface Pro. Presonus 192, DP88 (not currently used), RC500, ADL600, Focusrite ISA430, TC4000 reverb, SPL Phonitor, Monitoring: Event Opal, IK Multimedia MTM Studio Monitors x 5, DMAX audio Super Cubes. HD800s mastering reference headphones. Sequential Prophet 12, Prophet 12 modules x4, E-MU 4XT Ultra, Roland Fantom 8, Korg Pa3x, Roland Fp7-F, MOTU midi express 128, Xkey air 37, Studiolive CS18, Atari STE with Notator plus loads of microphones and plugins etc
User avatar
by admiralbumblebee on Mon Jul 13, 2020 2:54 pm
I went through and did more testing.

Windows - Fresh install, fresh installs of Studio One. Instances of Studio One were not run in their own systems, so they may have automatically imported settings from earlier versions.

  • V1 - Jitter, low latency.
  • V2 - Jitter depending on version.
  • V3.5.6 - No Jitter, longer latency.
  • V4 - No Jitter, longer latency.
  • V5 - No Jitter, longer latency.

Some notes:

  • V2 - Strange results. I tried multiple versions and there were different results.
  • V3.5.6 - Unable to try earlier versions. They all crashed on my system. I am confident that both behaviours exist during Version 3.
  • V4 - Something changed between this and my previous tests on Windows. No confidence in this result. According to my notes, V4 on Windows had the Jitter behaviour.
  • V5 - Not originally tested. I tested V5 in macOS, however there are Windows users that can replicate my results, so there must be a setting affecting this on Windows that I'm not familiar with?

macOS - Existing installs with default settings.
  • V4 - Jitter. Longer latency.
  • V5 - Jitter. Lower Latency.

Unable to run V3 and earlier without crashing. Probably catalina?

----------

Before writing the article I tested Versions the latest versions of 1, 2, 3 (unknown minor version) and the v2 version mentioned in Geert's article. This was done in Windows and they all had jitter.

I tested V4 and V5 on macOS and reliably reproduced the Jitter behaviour.

It might be worth me taking some time to explain the tradeoffs here (between extra buffer necessary to send offsets, and low-latency but 0-offset designs).

However I am still firmly in the camp that this is a bug for one specific reason (which I harp on constantly for other software's):

Any basic functionality with significant undocumented behaviour, even if intentional, is faulty from the user's POV. It's very simple to document the tradeoffs between these design choices, assuming they are, it could just as easily be a bug given the macOS V5 behaviour!

Additionally, the varied behaviour between Windows and macOS points towards this either being a bug or we are all missing a very important setting that affects this behaviour.

As usual, I strongly encourage anyone reading to try this themselves.

User avatar
by niles on Mon Jul 13, 2020 4:48 pm
admiralbumblebee wroteI went through and did more testing.

Windows - Fresh install, fresh installs of Studio One. Instances of Studio One were not run in their own systems, so they may have automatically imported settings from earlier versions.

  • V3.5.6 - No Jitter, longer latency.
  • V4 - No Jitter, longer latency.
  • V5 - No Jitter, longer latency.
I bet you tested with an equal or bigger buffer than 256 samples. Here's more on that: viewtopic.php?p=144701#p144701

OS: Windows 11 Pro | HW: Gigabyte Z690-UD-DDR4 • INTEL i7 12700K • 64GB • 3x EVO 860 • NVIDIA GT1030 (@WQHD) • RME AIO
User avatar
by markrobotham on Mon Jul 13, 2020 6:14 pm
Jemusic wrote
The real issue is how accurately a virtual instrument consistently plays back the recorded midi data and the resulting sound the virtual instrument accurately reflects that midi data. What we are seeing now is timing inconsistencies from the virtual instrument. eg If a series of very accurately placed midi notes eg every crotchet (either from an input such as the Admiral's video or a midi track that has quantised notes on it) the resultant virtual instrument output will fluctuate timing wise up to 15 mS differences or so between every crotchet. Some on the money, some early, some late etc..


I`m seeing this (or hearing should I say) in SO5 v4 was pretty tight for me. One reason I loved SO is the midi always seemed to be tight, but I did remix from scratch using version 5 and omnisphere was drifting in an out of time quite noticeably. I`m not talk latency or anything like that here, just pure playback of midi data. I know this used to be an issue when dropout protection was set to anything other than minimum. I always use minimum for that reason, but in V5 it`s not good, something has changed. I`ll try and sort out a video when I get a chance.

-------------------------------------------------------------------------------

    Windows 10 Enterprise 21H1
    ASUS TUF B450-PRO GAMING
    ASUS VP28U 4K Monitor
    AMD Ryzen 9 5900X 12 Core
    64GB DDR4 3200MHZ
    MSI GeForce RTX 3060 GAMING X 12G
    Zoom UAC-8 Audio Interface
    Studio One 6 Pro (Latest Version)
    Notion 6 (Latest Version)
    Sonar Platinum x64 Lifetime

[ YouTube ] [ Omnisphere Libraries ]

136 postsPage 3 of 7
1, 2, 3, 4, 5, 6, 7

Who is online

Users browsing this forum: AAV, chesli, Hommi and 64 guests