audio dropouts in Cross Pack

[djeksotik]

hi, im new here

i just got my Cross Pack and updated to 1.3.2.
i dont have a laptop yet so i borrowed one: Core 2 duo 2 Ghz, 2 GB RAM, windows vista. i use the control vinyls.

the problem is, the audio has dropouts and momentary distortions, not ALL he time, but quite consistently. especially when one deck is playing and im loading a song to the other one, but sometimes at other times too.

i do the calibration and everything that the quick start guide tells me to. i´m not a fan of vista, in fact this is my first experience with windows vista. first i thought the problem was vista, specifically vista using lots of memory. i closed all the programs and vista was still using almost 50% of the 2 GB RAM. then i disabled some of the services, like the infamous superFetch etc. restarted, still had the same problem with audio. then i checked the task manager while Cross was running, and discovered that the RAM usage was only about 1,2 GB and the processor wasnt struggling either.

anyway, i was planning to make a separate partition devoted to mixvibes only from the very start.. but still the current audio problem baffles me. can it really be just vista´s fault? because it sure LOOKS like it, although there is no clear evidence.


i have one more question.
im VERY pleased with the scrathing capabilities of Cross, but when I juggle, Cross tends to get lost during very quick pullbacks (with the vinyl), it seems the mistake is always one revolution of the record. for example lets say one record has played 3/4 revolution, and then i pull it back to the beginning real fast, but instead of going BACK 3/4 revolutions, Cross goes FORWARD 1/4 revolutions, ending up one revolution too far from the point i wanted it to play. sometimes the leap forward is greater, 1/2 revolution or so.
this means that the "sticker" on the record stays in the right place ("clock"wise).

is this normal? that Cross cant read VERY fast pulls longer than 1/2 revolution or so? or is this a problem with my setup or smth?

i have fiddled with the buffer setting but this seems to have no effect regarding these problems.

and another thing, in my Cross Pack there was only a ~10 page "quick start guide" but no real manual. and i can´t find any information about Cross on the web either, other than the same quick start guide. have i missed something, or does Cross really come with no manuals and detailed info?

[djeksotik]

ok,
so i tried again, put the buffer to 512 samples, and it now seemed to work fine. then put it to 256 samples, and got a few dropouts. it still happens mostly when loading a song to a player, ESPECIALLY during ananylizing. this last fact seems perfectly logical to me. and i think im right in assuming its windows vistas fault. aren´t i?
seems like vistas fault because eralier the dropouts occurred unsystematically too, there were clean periods and periods with some dropouts.
but it still seems weird to me that when the computer´s specs are above recommended specs, i have to use 512 buffer, even on vista. i have heard a lot of bad about vista, and to keep away from it (you all know it ), and i would definitely LIKE to blame it on vista... but still it feels a bit suspicious, doesnt it?

do you think when i install a fresh windows XP on a new partition, i shhouldnt have this problem? and i would like the buffer at least 256 samples if not lower, since 256 is already quite good for scratching.

otherwise im very satisfied with the product, though i live in estonia, so in the future i will have to ORDER control vinyls etc - this sucks, but otherwise great bang for the buck!

[phadedvision]

Hi eksotik,

Vista's definitely a good thing to avoid, especially since that's also eating up RAM. I'd recommend using XP or 7, but with 2gb of RAM, XP might be a better choice. I think the sticker drift you're getting has to do with the dropouts.

[djeksotik]

so youre saying it has to do with RAM? but how is it then that only 60% or RAM is used?

***
Recommanded Configuration:
Windows XP SP2 or Vista
Intel Core Duo 2, 2 GHz
1G RAM
***
is this a lie then? because like i said my specs are better than these, and it clearly says vista there. is the recommended specs meant for 512 and greater buffers? do i get low latencies only with the newest computers? if so i think i can deal with it, i would just like some information!


and the sticker drift, i wouldn´t call it drift, more like an abrupt jump. like i said the sticker actually stays in the right place, meaning when its supposed to be at 12 o´clock, it stays at 12 o´clock , just one revolution too far. seems like the anti-sticker-drift-system is causing this? looks to me its thinking the needle has skipped or smth, and its trying to "ignore" this "skipping".. ??


but i haven´t seen a video where Cross is being juggled on properly. the only one i found is this:

http://www.youtube.com/watch?v=3sdFr3EmEQg

its mostly just scratching (he does a short juggle at the very end, looks like he pulls the record back more than half a revolution, and it really comes back, unlike mine. but then again he isnt pulling it very fast) , and like i said, im very happy with scratching on Cross, just superb.

[Support@MixVibes]

and the sticker drift, i wouldn´t call it drift, more like an abrupt jump. like i said the sticker actually stays in the right place, meaning when its supposed to be at 12 o´clock, it stays at 12 o´clock , just one revolution too far. seems like the anti-sticker-drift-system is causing this? looks to me its thinking the needle has skipped or smth, and its trying to "ignore" this "skipping".. ??

I'm assuming you're in relative mode so, it's the anti-skip that is working.
What does it give in absolute mode ?

For all the other issues :
- Vista is not so good
- you can go down to 128 samples with such system
- all your dropouts show that something is wrong in your setup

[hemskoc]

do you think when i install a fresh windows XP on a new partition, i shhouldnt have this problem? and i would like the buffer at least 256 samples if not lower, since 256 is already quite good for scratching.
!

I run Cross on a Celeron M 1.4 (2.2ghz equivelent ) w 1 gig ram, and my sample rate is @ 256 / 48k sample rate..on big spin backs, i sometimes get slight drift, but it comes 'back'...and yes; i think its caused by sample rate/data packet loss issues from what i gather.
As 'Mr.Support' says, there is something conflicting in your system, as your system is powerful enough to crank the hell outa Cross!!!(wireless/cam/ext...).
Before you try a partition, try disabling as much as you can (unnecessary programs running in the background hogging resources).

I have Cross on a dedicated partition w/XP (2)version.
If the 'above' does not work for you, I would definitely try a partition of xp; and follow the guide in 'tips and tutorials' on setting up xp for MV here:
viewtopic.php?f=207&t=11532
You can also try changing your 'sample' rate for more accurate vinyl timecode reading (see this topic:)
viewtopic.php?f=207&t=16170

Good luck
Cheers

[Jim B]

Also you could download and run this useful tool

http://www.thesycon.de/deu/latency_check.shtml

It will show if you have any devices or drivers that may or may be not causing your drop outs.

Run the tool for a few minutes. You do not want to see red spikes.

[djeksotik]

thanks for your replies.

i wanted to make separate partition from the very beginning, i just didnt have a copy of win XP, and i wanted to start practicing with mixvibes right away, as i have live performances coming. and i was just curious if people have been known to run mixvibes on vista with no problems. but so much for vista, dont want to touch it again.

so now i have made a separate parition with XP. memory usage when no programs are running has dropped about 90% (although XP´s supposedly inaccurate about mem usage info and the vista partition was full of crap - still quite a difference between XP and vista!). i dropped the latency to 128 samples, and now i get dropouts momentarily just when loading the first song (of a session) into a player, after that there are no dropouts. so i guess the problem is solved by now, but is the first-song-lag a usual thing?

i haven´t tested juggling in absolute thoroughly yet, but the whole point of juggling in relative for me is to avoid needle skipping! my needles are quite old (they still sound good though ) and so on, the decks are not the best either. so its a bit of a dissapointment that i still have to deal with skipping when juggling in relative mode. it seems strange to me, because the whole point of relative mode is to avoid skipping (at least one major point). that the anti-skipping mechanism is actually causing virtual skips.

is there a precise limit of speed at which Cross can still read the timecode? i may be totally mistaken here, but i have a hard time understanding, why is it difficult for the program to distinguish between physical needle skipping (the timecode being cut off completely and then returning, but in the next/previous groove) and a very fast spinback (no abrupt cutoff, but a very fast speed of timecode playing ), as these thingh seem completely different to me. the only reason i can come up with besides some limitation of the software, is that the needle IS ACTUALLY physically skipping at some point of the spinback (although it should be in the air quite long then?), in which case it is my own fault.
this last theory is supported by the fact that this skipping seems to be occurring randomly: im doing some juggling pattern, and on the n-th repetition of the pattern the skip occurs, not every time.
(maybe i should try putting even more weight on the needles)

still: any comments on the software side?

i dont want to be pain in the ass with my detailed questions, i just want to know the limits of the program, and if im doing everything right on my part, and maybe even give some useful feedback.

would be cool if some juggler could give his opinion/experience on this matter.

how do i change the sample rate? because i cant click on 48 000 in the Cross preferences..

[hemskoc]

Hey how worn are your needles? a little or lots? What is the % average value you are getting of timecode read? less than 95%?
Sorry so many factors ( and questions)

Needles WILL make a difference,(especially if your juggling) and try weighting them down a little...have u got the counter weight just over-balanced, or really heavy? Im thinking your needles are 'in the air' as you put it for that 'blank' moment you are getting.
if im doing hard backspins with digital vinyl, i find i have to put more weight the needles (unlike real vinyl). Spose its something to do with the accuracy of the timecode read...
Might be the grooves are not as wide/deep maybe on timecode wax(someone correct me on this).

For the dropouts at the start of tracks, try changing the "cache value" on the first page of preferences to a smaller amount of memory usage. I did this and it solved my probs...
there is a thread about this issue also:
viewtopic.php?f=207&t=16177

As for the sample rate switchhing, someone else here will know how to change this ( i dont have a u46, i use edirol fa66 <goes up to 192k sample rate!>).
hope this helps
Cheers

[Support@MixVibes]

I made a simple test : a very long and fast back spin.

At a very very very high speed, yes the timecode is weird, but as far as the speed drop it is handled correctly.
So based on this, I don't think that on a full turn you can be quicker than a vinyl at full speed in backward.

Also to reach that speed, I have to put a lot of energy on a lot of turns.
What I mean is that you can't humanly give it an acceleration with such a raise on a single turn. BUT if you do, the strengh you need to put down is so high that YOU WILL have skip.