XITEd Forum for Sonic Core XITE-1 and Scope

I don't know where to post this, because there is a place for XITE's bug, but not the thing wich are working better...( sorry ADMIN )

we are few 3rd party to use a kind of c/c++ code inside our devices, GOST and DIGITALAUDIOSOFT use this technic on few devices.
Problems are that we need to "deal" with an important latency as soon as we want to work with audio, SCOPE->PC calculation + the return PC->SCOPE = MORE TIME...
...and this technic was just used for ASYNCHRONE calculations ( like for DAS MASTER IT EQ, DAS LEGEND, DAS C350 and GOST's devices,Thomas, if i'm wrong, i'll correct this sentence of course.)

EG : With the same device ( a FIR linear phase crossover ) the latency with SCOPE PRO and SFP4.5 is more or less 40ms.this is not important for a crossover (but for a realtime eq, it is not possible)
BUT NOW, with XITE-1, the latency with this device becomes 10ms ! 10 ms is more than real time of course, but it opens some BIG doors for new kind of devices...

so, i have to say, 10^31 "thank you" to SonicCore.

Eric

ps:i'd like to hear from SonicCore : "here is the SDK for XITE..."

Yes, we are all looking forward to more awesome (i.e. UAD-killer) devices!!!

I would recommend just sending SC a nice friendly email reminder and tell them you support them on releasing an SDK for Xite

Greg

p.s. SC is rarely, if ever on this forum

Hi Greg,

i've sent more than one mail to Ralf D,or Holger, and the last one, about SDK, is dated from 4th of july .
i always get answers from SC, but it seems that SDK is not for tommorow...it'll take time...but "taking time" doesn't mean "never"...
this the reason of my wish...

UAD KILLER DEVICES ? but what "UAD" is ?
they already exist ! eg : "GOST" devices and there is no FIR linear phase crossover on UAD isn't it ? there is also no plugin like DAS noise reducer (see the video on our website, we it made especially for Fairlight constellation console few years ago...) a clone of the most popular noise reducer hardware used by all movie sound engineer...http://www.cedar-audio.com/products/products.html#DNS3000

Eric

Ankyu...............

we use sys AND dll for sync/async and mididata from bool up to double;
no need to edit;

-----------------------------
Scope PRO:

fastest !!

4.44ms at 44100
4.08ms at 48000
2.04ms at 96000 (is about 0.7m distance!! if anyone get closer to you (s)he will get a kiss or a punch)
=196samples
---------------
maybe even faster only sc knows_

without any tricks;

5.85ms at 44100
5.37ms at 48000
2.69ms at 96000

=258samples at fastest ULLI

i would expect the pciexpr. to be faster;
on the other hand theres the cable between xite and pccard;
could you measure again please?


good vibes

Yes, so it should be easy to 'kill' UAD, since all they do is license the GUI's of famous boxes and make the end user pay the price.

Greg

Hi Thomas,

i talked about the last "wallX" and "DAS FIR Crossover". i'll make new measurement with a simple "SYNC ADD" it 'll be better for a basic test.
i don't know how to measure asynchrone...help is welcome

@Greg



UAD make very good plugins, that's true, but, for sure, the trademark+the so beautiful GUI help to sell more.
that's the first time i read a man writing what i/we think all.
Thanks to be honnest.

cheers,

Eric

@Jimmy,

i prefer the black version
http://www.digitalaudiosoft.com/plugins%20digitalaudiosoft/DAS%20SC_GATE%20BLACK.zip

Eric

SORRY ! i made a big mistake !!! my ulli were not set the same on both SFP !

so the result are the same ...i'm a bit sad.

ulli set to 4ms, fs = 44100
i send a dirac to wallX 255 nuttal asio left, dry asio right, recording to wavelab with asio, result = 12ms on SCP5 and on SFP 4.5 ... = no change !

i was wrong or too "x-ited" to work with XITED-1...

Eric

Eric you are too kind.
I have wanted that for a long time.
Just like I prefer my women too...........
Fat and Black....Ankyu.

i think i know all the values for scopeboards;
but i would be interested in a wider range of values for Xite;
specialy if different setups could make a change?
i dont think so but who knows if non measured?

so please check X ite1

here we go:

_empty project;
_load both devices in the project;
_assign them to the same dsp;
_connect HOST_LOOP out to CHECK in;
_connect CHECK out to HOST_LOOP in;
_open CHECK interface;
_open ULLI settings;
-----------

start with slowest ULLI
wait for the values to stabilieze
and write down the results for all ULLI settings;

a switch of the samplerate should make no difference!!

--------
here are the values to compare:
(if you have any other please tell me)

_ 258
_ 386
_ 642
_ 1154
_ 2178
SAMPLES for scopeboards
----------

thank you vibes


EDIT:

if one thinks to check any other device than HOST_LOOP
one should not get fooled by effects that modify the measurement signal and therefor modify the result;
but you may use as i do to measure all noneffected lines and cables of your studio;

phase aligned edit vibes from vienna

Brotha.Man tgstgs............
I have catalouged each step w/ a jpeg.
I might be needing some DSP files...?
I get no results so far, but would love to check it out.

Last time (long time ago) I checked it was 194 samples delay, but I might remember it wrong.

here are my test

and the 2 others

Eric

running at 44.1

"host loop" and "check" assigned to one dsp = 390s

"host loop" assigned to old dsp, and "check" to another old dsp = 390s

"host loop" assigned to old dsp, and "check" to a new one = 394s

conclusion : we need to take care to always assign to the same kind of dsp ( old to old, and new to new ).

Thomas, your "check" device is very important for us !
thank you, it is so fast...i did the same using wavelab,and i took more time...

Eric

my measurement were about SYS developement ( 4 samples delay btw old to new dsp ).
but but but !!!
the "dsp loop" and "check" assigned to one dsp (old or new ) = 0s ( of course ), but if i assign "dsp loop" to dsp 6 and "check" to dsp 16 or any others dsp, the result is still 0 sample !
seems they have solved phase alignement !
they are the best developpers in the dsp word !

thanks SC and thanks to Thomas ( GOST )

Eric

Eric...thats great as I know you are using the XITE-1.
Do you have DSP files to share, or is my set up configured wrong..?
Do I need to run Reaper or Cubase in the background as a host...........?
Whats the trick...?

Ankyu

Hi Jimmy,

dsp loop is just a in connected to out,so you can choose any "small device" and you bypass it. but this is only for DSP measurement.
to check a device, you do the same with the device you want.
you need to copy the .sys files, Thomas gave us, inside the sys folder of xited software, if scp5 is running, you close it before copying the files.after you launch scp5 and you do the same test as i did (jpeg).
how to read : 390 samples at 44.1 on SCP5 XITE-1 ( "host loop" -> "check" ) is the latency to go to the pc and to return from PC = 8 ms, because SYS use PC calculation. if you check a DSP device, the number is 0 samples for DAS2A or das1610 ! and for a simple eq it can be 24 samples...it can vary of course. so 390 = 195 samples...
warp69 wrote 194 samples (on scope ?), thomas 193 samples with a scope, the same result as me with scope 4.5 and scope pro.
with this FREWARE from Thomas, we can check if a mixer or any stereo device is well phased.you simply need to compare left to right.and it will help us to make big devices using old dsp with new dsp and .SYS if we take care of the 4 samples delay to be phased...to be more precise, i needed to add few sample delay btw my FIR lowpass and my fir hipass to be correctly phased.
as i wrote, you can do the same with wavelab or any audio editor, but it takes long time...and i don't have time to loose time is money ( i don't remember who said that ! ).
Eric

As usual I am a Pendajo........
I forgot that when using SYS. files that app needs to be re started.
So I had everything I needed, except I lost my memory/braincells from excessive tippers of Jager..

I have everything working now.
But since I am a moron with the science amd math, could you explain what I have in the screen above.

Thanks Again Brotha' Men Eric & tgstgs...

I use 7msec/44.1k live and have zero issues with any latency, so whats all of this stuff mean for a Schmuck like me...?

so, it looks good ( like mine but with the ully set to min ),you need to assign each module to the same dsp.
but what is 7ms/44.1 live ? i don't understand you are running at 96khz with 1 ms ully !
266s is the latency ( s = samples ) as you check with SYS files, you need to divide/2 the result = 133 samples latency.
tgstgs ( Thomas ) wrote 258 samples on his scope.
as you can see with my test, XITE as "always" few samples more latency than scope.

one thing,you don't need to put the right fader at max, let it at the middle.

ps : i like your "bleu note" module and the blue note, for a musician like me is something important, and it is also good for our (old) memory...we are not so young here

I am so old,.................................. when I go Antique shopping they close the door behind me.


OK, I understand now the prupose of this test.
I will re do it, but I have having some much fun patching Modular and creating synths and presets to gig with, the science and math is secondary.
As long as I press a note and hear it, I am not worried.
But its nice to know things are good I suppose.
I know from taking a test on stage once that a ride cymbal from 12 feet away takes 6msec's to reach my ear. Cool.
Know I know why I like the sound bouncing off of the back walls when I sit in the balconys' during local concerts and symphony performances...........

Orevua Mon Frere...........

thanx to all so far very interesting;

if i got it right the pccard has a 8sample buffer for the old and a 4 sample buffer for the new dsps!?

i would guess this the main trouble for the old devices to run;

we use realtime measurement in our devices to be sample exact no matter how much samples delay there are;

im climbing in the mountains of austria;
having troubles with my discs i do a break right now;
taking the next wall tomorrow;

to get old is no probl. you know
its the pain that makes it realy FU****g

good vibes from the alps

If you ever make it to the states I am well schooled with El Capitan and Half DOme.
I won't ever climb those but there are many, many more smaller Granite faces in the park
Especially by the Glacier.
Yosemite is so pretty in Sept/October beofre tha passes close.

Sorry for being offtopic but that is one AWESOME place..