Multitrack recording skipping

[robbocurry]

And the video too:

[MartinW]

Very interesting!

[debzdoodle]

This is all very interesting I have to admit. I feel so bad for those who have had nothing but problems with recording on the QU16. I myself have had great success but I would be very upset if I had gone through these issues. However, this new ‘suggestion’ has definitely got me thinking. I lay my hard drive under the console on foam which again could explain my lack of issues. This video does suggest vibration could be the key. Not sure why the A&H guys haven’t suggested this before???…..It could have saved a lot of frustration and in the case of Prodigio and maybe more, a move to something different. I understand that there are some who have checked out the vibration theory but are still having problems. But knowing about this vibration theory a few months back may have cured the problem for most users…..

[Anonymous]

Robocurry thanks for posting that video link, very interesting!! Until now I couldn’t lean toward whether it was a fault in the QU16 or the noise / vibration but this video has swayed my thinking on exactly what caused my (any many others here) skipping. I am now leaning toward loud sound from the front of house / vibration being the cause of my skipping as opposed to being any problem with the QU16 itself, and here’s why:

1. I noted from another thread in these forums that some users reported far less occurrences of the skipping than I experienced. Some guys said they’d get a skip every few min or so, my skipping occurred every 10 seconds or so, BUT ONLY WHILE THE BAND WAS PLAYING and NO SKIPPING AT ALL between songs while only the singer was talking to the crowd.

2. If it were a firmware problem, or any other sort of technical problem with the desk surely the symptoms would be more uniform across all users; some guys have never had the skipping problem, yet others using the same firmware version and approved hard drives ARE having the problems.

3. I have only recorded once from the QU16 to an external HD and the room we were in was a shocker for noise. If anyone here is from Sydney they might know the venue, the North Bondi RSL and it’s mostly glass and hard floors, a very noisy room. Add to that, we are an AC/DC Tribute show and we’re VERY loud through the FOH. The desk was at the back of the room, probably about 15m from the stage, and the volume would definitely have still been very high back there. The hard drive was sitting in a tray right behind the desk.

I won’t have an opportunity to do further testing until the next time we work with this particular desk which unfortunately isn’t until some time in June. But meanwhile I will be making up some sort of foam lined box to house the external hard drive which will insulate it from both volume and vibration. I will certainly report the results here, of course. And hopefully in the meantime someone else might be able to do some testing sooner and report their results.

[WaihekeSoundie]

Interestingly I found this on one of the RME forums regarding recording direct to disc from their Fireface UFX:
https://www.rme-audio.de/forum/viewtopic.php?id=15454

I have used the RME gear myself for years and found their drivers rock-solid.
The suggestion in this little blog post is to use SSD drives and I note that RME allow for USB sticks upto 2TB.
Why do A&H explicitly advise against USB sticks? Is a SATA SSD formatted to FAT32 any different to a FAT32 USB stick vs Disk??
Anyone tried an SSD?

How come this issue doesn’t occur with DAW- laptops have HDD right? Better buffering? Or does it happen infact..
I noticed in one these threads someone pointed out that you can stream USB B and write to the QDrive simultaneously and it does seem to work.. wondering if someone having these issues can try a test with USB B and QDrive simultaneously– see if it is a QDrive issue only.

Andy

[Anonymous]

Just finished listening to 4 sets from Saturday night

Not one skip, It appears that 1.4 has fixed it ( at least for me)

I reformatted the Hard Drive with the board at the beginning of the night
I used a WD My Passport, 500G
I placed it on top of my rack on it’s side, between the mixer and side of the rack ( This puts the Hard Drive about 18″ from my floor tom )
I stopped the recording after each set. After all was done I had a 4 folders with multitracks of each set on my hard drive

I don’t think vibration is a issue. I’ve used the same setup with a PreSonus SL 16.0.2 into a Mac Mini With the Hard drive in the top of the mixer stuck between the mixer and side of the rack and never had a problem. I Think it was just a firmware tweak that seems to be corrected now.

Before the update I would get about 10 skips per night ( 4-50 minute sets). They were quick skips, sometimes almost not noticeable unless you were listening for it.

I’ll record another gig in a few weeks to see if I get the same results

Ken

[WaihekeSoundie]

Reading the threads this is what I take away- that the 1.4 f/ware did address issues and improved things: some people with issues reported them fixed by 1.4 (perhaps buffering was changed or extended) but that given continuous vibration or perhaps specific modes of vibration, skipping will still occur if the conditions are right: Some users with issues reported them fixed at 1.4 only to find in high noise environments that they did intact still have issues.
Not all users will run in these environments however. Maybe most won’t in fact.

If this (speculation) is the case I’m very curious as to whether:
a) DAWs would suffer the same skips in the same environments- maybe laptops etc have better buffering;
b) Whether solid state drives for direct to disk recording are the answer..

Love my QU but I really think A&H should chip in on this and tell us what they know rather than leaving us to try and figure and test this out without background. It is affecting quite a few people– nothing worse than making a recording to find it is broken for reasons outside your control.

Andy

[robbocurry]

+1 on what you said Andy;)

[Anonymous]

Interesting video, and a surprising result for sure, especially in a noisy data centre environment too!

Just for sake of clarity on what I said earlier, yes vibration can cause issues reading and writing sectors on hard drives, but the buffer should be more than capable of handling the data whilst the drive sorts itself out. In 1.4 they’ve increased the buffer size which seems to have helped some people.

Don’t get me wrong though, I haven’t had the opportunity to sit and do any testing at all on this. I’m just going off the experiences others have had, and providing my input. My reasons for scepticism on the vibration theory, are that HDDs should be more than capable of handling vibration within reason, and it doesn’t seem to be affecting those using laptops or similar using the other port. Maybe it’s the computer itself absorbing this vibration, or perhaps the casing on these external drives are just not up to the task of isolating the drive well enough from it’s environment.

In theory, if vibration is the issue, the SSD drive should be flawless even if it’s placed on top of your kick drum (not recommended!), but SSDs can have their own issues if they’re not TRIM.

Bottom line is I agree that A&H should chip in and provide some insight into the issue, given they will probably have tested it to death!

[Nicola A&H]

Hi all,

Where to start.

Yes the buffer size is a key factor in preventing dropouts in the write process, as it compensates for the difference between the rate at which data is transmitted and the rate at which it can be written on the drive. When a USB storage device causes a pause that is longer than the buffer time, a dropout occurs and data is lost. Qu-Drive has a limited buffer when compared to a computer, but as some of you have noticed, one of the V1.4 improvements to Qu-Drive was allocating some extra memory for buffering.

As we cannot reproduce the dropouts with any of our recommended drives, freshly formatted and V1.4 running on the mixer, we can only assume most of the issues reported here are due to mechanical conditions such as vibration. Robbocurry and others have posted about this, and this is what I found on an HGST tech sheet (look at the test results of a hard drive on a shaker in their article!):

One of the greatest hindrances to hard disk performance is vibration. Like a needle on a record, the disk drive’s head must try to follow narrow data tracks in order to read (or write) information. Physical disturbances can throw the head off-track and cause a delay while the actuator repositions it. This eventually has an impact on the hard drive’s input/output performance.

You are also correct in thinking that SSD and flash thumb drives are virtually immune to this problem. However, SSD drives can be prone to other issues such as performance degradation over time as DMAudio has pointed out. As for USB keys, most aren’t fast enough for multitrack recording, hence our reluctance in recommending this medium. Qu-Drive requires a continuous data rate of around 21 Mbit/s when recording multitrack. Unfortunately manufacturers of USB keys typically publish burst (max) write transfer rates, which are of little significance here. Sustained rates are seldom published.

We had good reports of Sandisk Extreme USB keys (available in 16, 32 and 64 GB) and must say this is looking promising here. We need some further testing but we definitely want to include some of these devices in our list of recommended peripherals. However please be aware that another problem common to both SSD and USB keys is the batch variance of memory chipsets used by manufacturers – a device might perform well in our tests, but the same model purchased in the US or six months down the line could show different performance. I guess this is the reason why other pro audio manufacturers have been much vaguer with their recommendations.

Ultimately our list should be taken as a guideline only and you should always test the USB device thoroughly before putting into service.

Hope this helps.

[Anonymous]

Great response Nicola, thanks so much for taking the time to lay it all out for us.

Seems like a USB SSD drive with TRIM could be the safest solution, and with prices coming down (for the moment), we may see larger drives for larger sessions.

Hopefully the USB sticks will continue to improve, so even a bad one will still meet the transfer rates.

[WaihekeSoundie]

Many thanks Nicola.. very helpful.
Those USB 3.0 extreme USB sticks are reputedly very fast– 170mbit/sec over USB 3.. over USB 2..??
But 64GB would give around 5-6hrs of QU16 multi track record time I believe… prob worth a go then

Thanks Nicola

Andy

[robbocurry]

Hey Andy,
I’ve used a 16gb Sandisk USB3 stick and it worked great on the short 8 channel recordings I made – very fast to arm too.
Lots of song length recordings with no trouble at all. Will give it a go with all channels next week before I hop to a 64gb version.
Just want to echo what you and DMAudio have said and thank Nicola for his clarification of the issue – ta Nicola:)

[Davec]

I recorded a show yesterday, two sessions, one of 2 1/2 hours and one just over an hour. I noticed at the end of the first session as I was stopping the recording that the elapsed time seemed to be hiccuping. On transferring it to my PC it wouldn’t transfer track 1. Somehow it got corrupted. I’ve copied it over with a file repair software and it seems ok, although there was only a few minutes actually used on that channel.
Could be the HD packing up

[Davec]

Now I’m getting skipping! Towards the end of a 2&1/2 hour recording on a Seagate Free agent 500g. I’ve dozen 8 or 9 trouble free sessions, including ones this long. Firmware 1.4.

[Author]

Posts