Mailist to discuss all issues regarding the Digitech GSP-2101 Guitar FX Processor

Hi,
  I was wondering,can you change the crystal and make the 2101 
faster?I bet there's more to it than that but whado I know?:)
Thanks for reading!
                    Paul

Which crystal the DSP one or the Zilog? I would leave them, the 10 
MHz Zilog models have a 20 MHz crystal (they divide by 2 internally). 
There are also timer and interrupt issues that are in the firmware, 
so if it aint broke don't fix it! Absolutely DO NOT change the DSP 
crystal speed!

Only the older Zilog mcu can be upgraded to higher speed, but also 
involves a firmware change (to adjust for timers and etc having a 
different clock speed)

Just my 2 cents since I do embedded work for a living.

--- In gsp-2101@yahoogroups.com, "paulscary1" <PAULSCARY@M...> wrote:

> 
> Hi,
>   I was wondering,can you change the crystal and make the 2101 
> faster?I bet there's more to it than that but whado I know?:)
> Thanks for reading!
>                     Paul

Hi,
  Didn't even see the DSP crystal at first.It's hiding under the 
  Expansion board.The Zilog mcu reads:

                z8018010vsc
                zi80  mpu
                9407     6n
  Don't know if this is upgradeable.Is it a common part or native
  to Digitech?Funny...the upgrade chips read v3.00 but on boot it
  displays ver. 3.00.00a.My knowledge of this stuff is pretty basic.
  It's just that it seems that it's...well...slow.Nevermind using
  the FC to change multiple parameters being slow,but even with no
  effects(not bypassed)it "feels" a TINY bit slow.I thought I was
  imagining it until a friend of mine,who is a faster more technical
  player than myself,tried it for the first time,played some stuff
  and said "Is there a delay on this or something?".Don't get me 
  wrong.I love this unit...more than later versions on the merits
  of it's interface,crazy routing and control options,etc.Thanks
  for your response,Mark!
                  Paul

  
--- In gsp-2101@yahoogroups.com, "Mark" <mark@w...> wrote:
> 
> Which crystal the DSP one or the Zilog? I would leave them, the 10 
> MHz Zilog models have a 20 MHz crystal (they divide by 2 
internally). 
> There are also timer and interrupt issues that are in the 
firmware, 
> so if it aint broke don't fix it! Absolutely DO NOT change the DSP 
> crystal speed!
> 
> Only the older Zilog mcu can be upgraded to higher speed, but also 
> involves a firmware change (to adjust for timers and etc having a 
> different clock speed)
> 
> Just my 2 cents since I do embedded work for a living.
> 
> --- In gsp-2101@yahoogroups.com, "paulscary1" <PAULSCARY@M...> 
wrote:

> > 
> > Hi,
> >   I was wondering,can you change the crystal and make the 2101 
> > faster?I bet there's more to it than that but whado I know?:)
> > Thanks for reading!
> >                     Paul

Ooooops.  Ignore my previous post.  Since your display reads 3.00.00.a, the 
crystal upgrade has already been done.  don't do anything.

At 10:31 PM 11/17/2004 +0000, you wrote:

>Hi,
>   Didn't even see the DSP crystal at first.It's hiding under the
>   Expansion board.The Zilog mcu reads:
>
>                 z8018010vsc
>                 zi80  mpu
>                 9407     6n
>   Don't know if this is upgradeable.Is it a common part or native
>   to Digitech?Funny...the upgrade chips read v3.00 but on boot it
>   displays ver. 3.00.00a.My knowledge of this stuff is pretty basic.
>   It's just that it seems that it's...well...slow.Nevermind using
>   the FC to change multiple parameters being slow,but even with no
>   effects(not bypassed)it "feels" a TINY bit slow.I thought I was
>   imagining it until a friend of mine,who is a faster more technical
>   player than myself,tried it for the first time,played some stuff
>   and said "Is there a delay on this or something?".Don't get me
>   wrong.I love this unit...more than later versions on the merits
>   of it's interface,crazy routing and control options,etc.Thanks
>   for your response,Mark!
>                   Paul
>
>
>--- In gsp-2101@yahoogroups.com, "Mark" <mark@w...> wrote:
> >
> > Which crystal the DSP one or the Zilog? I would leave them, the 10
> > MHz Zilog models have a 20 MHz crystal (they divide by 2
>internally).
> > There are also timer and interrupt issues that are in the
>firmware,
> > so if it aint broke don't fix it! Absolutely DO NOT change the DSP
> > crystal speed!
> >
> > Only the older Zilog mcu can be upgraded to higher speed, but also
> > involves a firmware change (to adjust for timers and etc having a
> > different clock speed)
> >
> > Just my 2 cents since I do embedded work for a living.
> >
> > --- In gsp-2101@yahoogroups.com, "paulscary1" <PAULSCARY@M...>
>wrote:
> > >
> > > Hi,
> > >   I was wondering,can you change the crystal and make the 2101
> > > faster?I bet there's more to it than that but whado I know?:)
> > > Thanks for reading!
> > >                     Paul
>
>
>
>
>Yahoo! Groups 
>Sponsor<http://us.ard.yahoo.com/SIG=129oher92/M=296572.5585671.6651487.3001176/D=grplch/S=1705031959:HM/EXP=1100817098/A=2343726/R=0/SIG=12ifu4te1/*http://clk.atdmt.com/VON/go/yhxxxvon01900091von/direct/01/&time=1100730698641543> 
>
><http://us.ard.yahoo.com/SIG=129oher92/M=296572.5585671.6651487.3001176/D=grplch/S=1705031959:HM/EXP=1100817098/A=2343726/R=1/SIG=12ifu4te1/*http://clk.atdmt.com/VON/go/yhxxxvon01900091von/direct/01/&time=1100730698641543>
>Get unlimited calls to
>
>U.S./Canada
>
>
>----------
>Yahoo! Groups Links
>    * To visit your group on the web, go to:
>    * 
> <http://groups.yahoo.com/group/gsp-2101/>http://groups.yahoo.com/group/gsp-2101/ 
>
>    *
>    * To unsubscribe from this group, send an email to:
>    * 
> <mailto:gsp-2101-unsubscribe@yahoogroups.com?subject=Unsubscribe>gsp-2101-unsubscribe@yahoogroups.com 
>
>    *
>    * Your use of Yahoo! Groups is subject to the 
> <http://docs.yahoo.com/info/terms/>Yahoo! Terms of Service.

For future reference for anyone considering the crystal change, here are 
the part number from Mouser electronics, who will ship tiny orders like this:

Crystal: 559-FOX200-20            $0.47 each
Capacitors: 140-500N5-150J      $0.08 each

-Bob

Thanks for all your help,everybody!would like to see a reissue of 
this unit that ran a little quicker!I like the way the display and 
paging works better than later models.Thanks again!
                                         Paul

I did a little test. Out of curiosity, I recorded my GSP2101 to my 
computer using Cakewalk, with the effects send going to one track and 
the right output (with the wet/dry mix set to 100% dry) going to the 
other track. When I analyzed the wave data, the right output was 
around 6-7 ms delayed compared to the effects send. Since sound 
travels at around 1 foot per second, it should sort of sound like 
playing 6 feet (2 metres) farther from you're speakers. My 2101 is 
still at 10MHz though. I don't know if going to 20MHz would make a 
difference here.

--- In gsp-2101@yahoogroups.com, "paulscary1" <PAULSCARY@M...> wrote:

> 
> Hi,
>   Didn't even see the DSP crystal at first.It's hiding under the 
>   Expansion board.The Zilog mcu reads:
> 
>                 z8018010vsc
>                 zi80  mpu
>                 9407     6n
>   Don't know if this is upgradeable.Is it a common part or native
>   to Digitech?Funny...the upgrade chips read v3.00 but on boot it
>   displays ver. 3.00.00a.My knowledge of this stuff is pretty basic.
>   It's just that it seems that it's...well...slow.Nevermind using
>   the FC to change multiple parameters being slow,but even with no
>   effects(not bypassed)it "feels" a TINY bit slow.I thought I was
>   imagining it until a friend of mine,who is a faster more technical
>   player than myself,tried it for the first time,played some stuff
>   and said "Is there a delay on this or something?".Don't get me 
>   wrong.I love this unit...more than later versions on the merits
>   of it's interface,crazy routing and control options,etc.Thanks
>   for your response,Mark!
>                   Paul

The 10/20 Mhz refers to the main Z80 processor that handles program 
switching, parameter control, MIDI I/O. A faster processor means more 
responsive footcontrol response, quicker program switching, etc, but 
does not affect the sound quality in any way.

Ths S-DISC chip(s) on the 2101 both run at 24.576 MHz (512 x sample 
rate) no matter what  the control processor runs at, so the DSP 
processing is the same speed.

Adrian

On Dec 6, 2004, at 7:00 PM, saber540ltd wrote:

>
>  I did a little test. Out of curiosity, I recorded my GSP2101 to my
>  computer using Cakewalk, with the effects send going to one track and
>  the right output (with the wet/dry mix set to 100% dry) going to the
>  other track. When I analyzed the wave data, the right output was
>  around 6-7 ms delayed compared to the effects send. Since sound
>  travels at around 1 foot per second, it should sort of sound like
>  playing 6 feet (2 metres) farther from you're speakers. My 2101 is
>  still at 10MHz though. I don't know if going to 20MHz would make a
>  difference here.
>
>  --- In gsp-2101@yahoogroups.com, "paulscary1" <PAULSCARY@M...> wrote:
>  >
>  > Hi,
>  >   Didn't even see the DSP crystal at first.It's hiding under the
>  >   Expansion board.The Zilog mcu reads:
>  >
>  >                 z8018010vsc
>  >                 zi80  mpu
>  >                 9407     6n
>  >   Don't know if this is upgradeable.Is it a common part or native
>  >   to Digitech?Funny...the upgrade chips read v3.00 but on boot it
>  >   displays ver. 3.00.00a.My knowledge of this stuff is pretty basic.
>  >   It's just that it seems that it's...well...slow.Nevermind using
>  >   the FC to change multiple parameters being slow,but even with no
>  >   effects(not bypassed)it "feels" a TINY bit slow.I thought I was
>  >   imagining it until a friend of mine,who is a faster more technical
>  >   player than myself,tried it for the first time,played some stuff
>  >   and said "Is there a delay on this or something?".Don't get me
>  >   wrong.I love this unit...more than later versions on the merits
>  >   of it's interface,crazy routing and control options,etc.Thanks
>  >   for your response,Mark!
>  >                   Paul
>
>
>
>
>
>
>
> Yahoo! Groups Sponsor
>
> ADVERTISEMENT
> <122603_LREC_bkup_BR_Ashanti.jpg>
> <l.gif>
>
> Yahoo! Groups Links
>
> 	• 	To visit your group on the web, go to:
> http://groups.yahoo.com/group/gsp-2101/
>  
> 	• 	 To unsubscribe from this group, send an email to:
> gsp-2101-unsubscribe@yahoogroups.com
>  
> 	• 	 Your use of Yahoo! Groups is subject to the Yahoo! Terms of 
> Service.
>
>

Thanks a lot for that info!


--- In gsp-2101@yahoogroups.com, Adrian Teo <adrian@f...> wrote:
> The 10/20 Mhz refers to the main Z80 processor that handles program 
> switching, parameter control, MIDI I/O. A faster processor means 
more 
> responsive footcontrol response, quicker program switching, etc, 
but 
> does not affect the sound quality in any way.
> 
> Ths S-DISC chip(s) on the 2101 both run at 24.576 MHz (512 x sample 
> rate) no matter what  the control processor runs at, so the DSP 
> processing is the same speed.
> 
> Adrian
> 
> On Dec 6, 2004, at 7:00 PM, saber540ltd wrote:
> 
> >
> >  I did a little test. Out of curiosity, I recorded my GSP2101 to 
my
> >  computer using Cakewalk, with the effects send going to one 
track and
> >  the right output (with the wet/dry mix set to 100% dry) going to 
the
> >  other track. When I analyzed the wave data, the right output was
> >  around 6-7 ms delayed compared to the effects send. Since sound
> >  travels at around 1 foot per second, it should sort of sound like
> >  playing 6 feet (2 metres) farther from you're speakers. My 2101 
is
> >  still at 10MHz though. I don't know if going to 20MHz would make 
a
> >  difference here.

Actually, I meant to write "1 foot per millisecond". Thanks for 
pointing out the flub Graham. Unfortunately, I deleted your e-mail 
after reading only the title, thinking it would be posted in the 
forum. If there was anything else, please send it again. Sorry, just 
another one of my flubs!

--- In gsp-2101@yahoogroups.com, "saber540ltd" <saber540ltd@y...> 
wrote:
When I analyzed the wave data, the right output was 

> around 6-7 ms delayed compared to the effects send. Since sound 
> travels at around 1 foot per second, it should sort of sound like 
> playing 6 feet (2 metres) farther from you're speakers.

Saber,

Just to clarify, sound speed is about 340 meters/second or 1122 
feet/second. If it was a foot/second you would hear music a minute 
after it was played at a concert room or stadium, VERY BORING I 
GUESS!!!!!

That delay you're mentioning is related to the time the analog 
signal goes through the A/D converters, then the microprocessor to 
apply all the effect algorithms and then out the D/A converters. In 
more contemporary units, with much faster data "bus" and 
microprocessors, you would noticed less of a delay.

If you take a look at the GSP-2101 manual, it actually warns you 
about this fact because if you mix the processed signal with a 100% 
dry one you might perceive some signal degradation or artifacts due 
to what is called phase-cancelling. What happens is, since the two  
sound waves have different "phase synchronization" (in other words 
they start at different points in time), the overlap of their out-of-
sync phases produces them to partially cancel out on those points in 
time where there is an amplitude difference between the two. May 
sound a bit technical, but empirically in the end, what you hear is 
a somehow noticeable sound degradation to a higher or lesser degree.

Therefore, what Digitech advices us not to do is mix a pure dry 
signal coming from the analog section or external FX loop with that 
coming from the internal digital section, as phase cancellation may 
occur.

Why would you want to mix both a pure analog dry with the internal 
processed signal? Most of us using an external distortion unit 
(especially if it's tube based) have tried at some point, intending 
on obtaining a more "warmish and rawish" sound, to keep some of the 
un-processed signal coming out unaltered. Well, Digitech was kind 
enough as to forewarn users about what may happen should you try 
this.

Hope this helps.

Gus.


--- In gsp-2101@yahoogroups.com, "saber540ltd" <saber540ltd@y...> 
wrote:
> 
> I did a little test. Out of curiosity, I recorded my GSP2101 to my 
> computer using Cakewalk, with the effects send going to one track 
and 
> the right output (with the wet/dry mix set to 100% dry) going to 
the 

> other track. When I analyzed the wave data, the right output was 
> around 6-7 ms delayed compared to the effects send. Since sound 
> travels at around 1 foot per second, it should sort of sound like 
> playing 6 feet (2 metres) farther from you're speakers. My 2101 is 
> still at 10MHz though. I don't know if going to 20MHz would make a 
> difference here.
>

I was reading your mind!!!

--- In gsp-2101@yahoogroups.com, "saber540ltd" <saber540ltd@y...> 
wrote:
> 
> Actually, I meant to write "1 foot per millisecond". Thanks for 
> pointing out the flub Graham. Unfortunately, I deleted your e-mail 
> after reading only the title, thinking it would be posted in the 
> forum. If there was anything else, please send it again. Sorry, 
just 
> another one of my flubs!
> 
> --- In gsp-2101@yahoogroups.com, "saber540ltd" <saber540ltd@y...> 
> wrote:
> When I analyzed the wave data, the right output was 
> > around 6-7 ms delayed compared to the effects send. Since sound 
> > travels at around 1 foot per second, it should sort of sound 
like 
> > playing 6 feet (2 metres) farther from you're speakers.

Gus,

I'm aware that summing a delayed signal with the original one will 
cause some harmonics to cancel out and some to be reinforced, like a 
phaser would do with zero modulation. I was only confirming in 
practice what paulscary1 was experiencing. I just thought it would be 
interesting to point out that even with the Mix set to 100% Dry, the 
signal goes through digital processing. But thanks for the additional 
info; I've had the 2101 for around 10 years and didn't remember 
reading that stuff in the manual.

The delay has never really bothered me but I remember reading of 
someone who prefered using external mixing of wet/dry because of the 
delay (I think it was on tubefreak.com). And after reading 
paulscary1's post, I was curious about the actual facts. 


--- In gsp-2101@yahoogroups.com, "gusfmm" <gusfmm@y...> wrote:
> 
> Saber,
> 
> Just to clarify, sound speed is about 340 meters/second or 1122 
> feet/second. If it was a foot/second you would hear music a minute 
> after it was played at a concert room or stadium, VERY BORING I 
> GUESS!!!!!
> 
> That delay you're mentioning is related to the time the analog 
> signal goes through the A/D converters, then the microprocessor to 
> apply all the effect algorithms and then out the D/A converters. In 
> more contemporary units, with much faster data "bus" and 
> microprocessors, you would noticed less of a delay.
> 
> If you take a look at the GSP-2101 manual, it actually warns you 
> about this fact because if you mix the processed signal with a 100% 
> dry one you might perceive some signal degradation or artifacts due 
> to what is called phase-cancelling. What happens is, since the two  
> sound waves have different "phase synchronization" (in other words 
> they start at different points in time), the overlap of their out-
of-
> sync phases produces them to partially cancel out on those points 
in 

> time where there is an amplitude difference between the two. May 
> sound a bit technical, but empirically in the end, what you hear is 
> a somehow noticeable sound degradation to a higher or lesser degree.
> 
> Therefore, what Digitech advices us not to do is mix a pure dry 
> signal coming from the analog section or external FX loop with that 
> coming from the internal digital section, as phase cancellation may 
> occur.
> 
> Why would you want to mix both a pure analog dry with the internal 
> processed signal? Most of us using an external distortion unit 
> (especially if it's tube based) have tried at some point, intending 
> on obtaining a more "warmish and rawish" sound, to keep some of the 
> un-processed signal coming out unaltered. Well, Digitech was kind 
> enough as to forewarn users about what may happen should you try 
> this.
> 
> Hope this helps.
> 
> Gus.

I apologize, I misunderstood a bit your trial. I just re-read it and 
see what you were saying. Still, what I said is somehow applicable 
in this case. If you look at page 103 in the manual, the FX loop's 
send coming from the analog section doesn't get "digitalized" 
whereas the rest of the signal paths go into the A/D/A.

How was the audible signal quality in one case and the other? Any 
noticeable difference?

Now, I also thought paulscary1 was more complaining about switching 
presets or responding to MIDI control messages or something. 
Otherwise, how would his friend notice that there was some delay?

Good experience Sabre!

Gus.

--- In gsp-2101@yahoogroups.com, "saber540ltd" <saber540ltd@y...> 
wrote:
> 
> Gus,
> 
> I'm aware that summing a delayed signal with the original one will 
> cause some harmonics to cancel out and some to be reinforced, like 
a 
> phaser would do with zero modulation. I was only confirming in 
> practice what paulscary1 was experiencing. I just thought it would 
be 
> interesting to point out that even with the Mix set to 100% Dry, 
the 
> signal goes through digital processing. But thanks for the 
additional 
> info; I've had the 2101 for around 10 years and didn't remember 
> reading that stuff in the manual.
> 
> The delay has never really bothered me but I remember reading of 
> someone who prefered using external mixing of wet/dry because of 
the 

> delay (I think it was on tubefreak.com). And after reading 
> paulscary1's post, I was curious about the actual facts. 
>

Gus,

According to the block diagram in the manual, you're absolutely 
right. I still wanted to see for myself exactly how long the delay 
was though, just out of curiosity.

When I first did the test I recorded the signals direct to computer 
without cabinet emulation. I just wanted to measure the delay so I 
used the clean tube setting with very low distortion and hit a short 
note, so I wasn't concerned with sound quality. I normally play 
through a power amp and speakers, and mic that when I want to record. 
But I did go back today and recorded a small sample after placing the 
global EQ flat to compare Effects Send with the main output at 100% 
dry. Though the slightly different playing in the two licks makes it 
harder to judge, I don't hear much difference in sound quality. It 
was played through an ADA MicroFet poweramp and speakers and recorded 
with an SM57. The first lick is through the Effects Send and the 
second is through the 1/4" Right Main output. Here's an mp3 of the 
clip so you can judge for yourself:
http://www.fullservesite.com/saber3/auxproc1.mp3

Regards


--- In gsp-2101@yahoogroups.com, "gusfmm" <gusfmm@y...> wrote:
> 
> I apologize, I misunderstood a bit your trial. I just re-read it 
and 

> see what you were saying. Still, what I said is somehow applicable 
> in this case. If you look at page 103 in the manual, the FX loop's 
> send coming from the analog section doesn't get "digitalized" 
> whereas the rest of the signal paths go into the A/D/A.
> 
> How was the audible signal quality in one case and the other? Any 
> noticeable difference?
> 
> Now, I also thought paulscary1 was more complaining about switching 
> presets or responding to MIDI control messages or something. 
> Otherwise, how would his friend notice that there was some delay?
> 
> Good experience Sabre!
> 
> Gus.