Skip Navigation.
Section 0

ChromaTalk Archives: October 2003

Chroma sound files in midi format ?

Jim Russo [21030433] · Sat, 04 Oct 2003 08:32:27 -0700

Greetings group,

I have a Chroma that has just been revived. Unfortunetly it lost it's program memory. The only way I have of loading sounds into it is Logic Audio on mac. Now unless I am missing something, the patch downloads on the site are all sys-ex files and Logic does not seem to recognize these. Has anyone converted them to midi files? or is there a way to load the sysex data into Logic ? Any help is greatly appreciated.

David Clarke [21030085++] · Sat, 04 Oct 2003 11:02:45 -0400

Some of the other Mac users might be able to provide some specific recommendations for sysex (Chris, can you help here?) - but if worse comes to worse, you should be able to reload the Chroma via the tape interface (i.e., dump an audio file to the TAPE IN port on the Chroma).

Chris Ryan [21030691] · Sat, 4 Oct 2003 13:25:34 -0700

There's another solution: dedicated sysex apps. For the Mac, there are two popular alternatives, depending on operating system:

See Editors & Librarians: Mac OS X for other alternatives and the latest info.

I use Logic Gold (6.0), but I haven't tried using it to send sysex. I can look it up in the manual (though I still have a 4.x manual with addendums--I was talking to Emagic/Apple just yesterday and it seems they still don't have a 6.x manual for Gold). Jim, what version of Logic are you using?

Jim Russo [21030433] · Sat, 04 Oct 2003 22:09:10 -0700

Chris Im using Logic Mac 4.7

We continued this discussion off the list; I pointed Jim to the Logic User Group [since moved from emagic.de to its own domain following the Apple Emagic acquisition] and the latest Logic updates.

Should 'spotted' Chromas go into the Registry?

David Clarke [21030085++] · Sat, 04 Oct 2003 11:39:45 -0400

All - if a Chroma is spotted somewhere (such as for auction on Ebay, where the serial number is clearly visible), do you think we should make an entry in the registry for it - or should the mechanism to get into the registry only be a submission from the owner of the Chroma itself?

Thoughts? Comments?

Luca Pilla [21010218] · Sat, 04 Oct 2003 17:51:48 +0200

Yes, every Chroma has rights to enter in database. I will write serial numer and Ebay

Chris Ryan [21030691] · Sat, 4 Oct 2003 13:18:57 -0700

I've added eBay items to the registry a couple of times recently. If you would like to send them on to me (past or future auctions), that would be great.

I still do hear from Chroma owners either who haven't seen the site before or haven't e-mailed previously; or just recently acquired a Chroma.

Jeffrey D. McEachin [21030073+] · Sat, 04 Oct 2003 09:54:22 -0700

All - if a Chroma is spotted somewhere (such as for auction on Ebay, where the serial number is clearly visible), do you think we should make an entry in the registry for it

Yes, it's a matter of public record at that point.

First Chroma Serial/Model Number Overlap?

See How many Chromas were manufactured? in the FAQ for links to related threads.

David Clarke [21030085++] · Sat, 04 Oct 2003 17:24:50 -0400

Up to now, all Chroma serial numbers have been unique - even when you look at them across model numbers.

For instance, if the last 4 digits of a serial number were "abcd", then there would only be one model keyboard (model 2101, 2102, 2103 or 2104) with those numbers.

Looking at the current registry there now looks to be two different models with the same serial. If there's isn't a typo, this may impact on how many Chroma's we think there may really be out there.

The current overlap is between S/N 21010218 (Luca Pilla) and S/N 21030218 (Paul Tillotson).

Paul/Luca, can you please check the S/N information from the tag on the back of your units to confirm the serial number information from the registry? (If they are both legimatly 0218, then there may have actually be more Chromas then what we thought...)

Paul Tillotson [21030218] · Sat, 4 Oct 2003 20:50:34 -0700

Ummm, no sorry I cannot. It's in Backstreet Music in Vancouver for repairs.

Wish I could.

Chris Ryan [21030691] · Sun, 5 Oct 2003 23:28:53 -0700

The current overlap is between S/N 21010218 (Luca Pilla) and S/N 21030218 (Paul Tillotson).

I can, at least, confirm that it's not my own typo.

Luca Pilla [21010218] · Mon, 06 Oct 2003 13:08:26 +0200

My Chroma doesn't have any tag, only a serial number near the support for voice cards. I attached the photo. Where this tag should be? There is something strange, don't you?

David Clarke [21030085++] · Mon, 06 Oct 2003 18:56:30 -0400

Most Chromas have their serial number marked in two locations.

The first is on a foil-type tag mounted on the power-supply/rear of the keyboard. (I have attached a representative picture here - the foil tag is just to the left of the orange/yellow circle.)

The second is typically in pen or marker on the inside of the keyboard itself. It is usually marked on the wood base, near the voice-cards.

Having the two locations is sort of a nice way to be able to confirm that the power supply/serial number actually matches the rest of the keyboard. (For instance, I have a Chroma Expander that actually has a 'regular' Chroma power supply - and serial number - installed. In that case, I have an Expander serial number on the inside, and a Chroma serial number on the outside. I'm sure this keyboard wasn't built this way - I know this mixup was a result of a past visit to a repair shop with a former owner).

David Clarke [21030085++] · Mon, 06 Oct 2003 19:18:51 -0400

Paul Tillotson wrote:

Ummm, no sorry I cannot. It's in Backstreet Music in Vancouver for repairs. Wish I could.

This might be a touchy question (depending on how long the keyboard has already been in the shop :-> ) - but do you have any estimate as to when you're expecting it back in your hands (so we could confirm the S/N)?

Paul Tillotson [21030218] · Mon, 6 Oct 2003 17:12:27 -0700

It's certainly not a touchy subject with me at this point. I'm used to dropping off my gear and not seeing it for many months. It was just dropped off one week ago, I'll start getting peevish about it in a year and a half. I would be surprised if it were back home before the end of November.

I'll try to remember to ask the serial number the next time I communicate with the tech. Did the serial number come from me or was it transfered from the seller's information on Chroma Talk?

Chris Ryan [21030691] · Mon, 6 Oct 2003 17:53:20 -0700

On Monday, October 6, 2003, at 05:12 PM, Paul Tillotson wrote:

Did the serial number come from me or was it transfered from the seller's information on Chroma Talk?

From you:

On Monday, June 3, 2002, at 10:06 AM, Paul Tillotson wrote:

Please welcome me to the fold at long last. I'm resting up after a marathon drive from Seattle WA to Bend, OR and back, where I picked up a beautifully preserved #21030218.

Memory problem fixed

Luca Pilla [21010218] · Thu, 09 Oct 2003 00:34:24 +0200

Hi!

I solved my first problem with memory chip and battery circuit, with a cable between the point B and C [see My Chroma is not happy, September 2003]. Now the RAM works fine. For your reference: I have had a lot of trouble in loading a bank, with this fix my Chroma loads an entire bank without problems. Now I have to fix the voice cards... thank you for your help

Fixing voice card

Luca Pilla [21010218] · Fri, 10 Oct 2003 20:55:39 +0200

See also Fixing voice cards (September 2003)

Hi all,

I'm trying to fix the voice cards (4!). I started with Set Split 31, then I started the alternate trim procedure and I discovered that the first not working voice card has one oscillator (from TP2) not soundig, except for some noise when I move the trim, and the other oscillator (from TP1) is sounding fine. In another voice card both the output from TP 1 and 2 are too low, not an octave higher as espected from the procedure like working voice cards. I swapped the voice card, but the results are the same for the not working voice cards. I measured the voltage from chip with my new digital voltmeter, and I found 4,8V in every voice cards. Where should I search for fixing? How can I check if a vco is bad?

Thank you

David Clarke [21030085++] · Sat, 11 Oct 2003 12:02:08 -0400

I'm trying to fix the voice cards (4!). I started with Set Split 31, then I started the alternate trim procedure and I discovered that the first not working voice card has one oscillator (from TP2) not soundig, except for some noise when I move the trim, and the other oscillator (from TP1) is sounding fine.

Luca - I would suggest that this is the best card to work on first, as the problem seems to be in the 'first part' of the channel circuitry (i.e., before the VCF and VCA).

The first thing to check is whether or not the necessary control voltage is making it succesfully into the board. The key values you'll be looking at are the "Pitch A" and "Pitch B" control voltages.

First, take a look at the voltage on Z21 pin 7. This is the Pitch control voltage for VCO A.

Now, take a look at Z23, pin 1. This is the Pitch control voltage for VCO B. If the VCO B control voltage isn't similar to that for VCO A, check the actual buffer stage (Z24). Specifically, look at Z23 pin 3. If Z23 pin 3 looks OK, but Z23 pin doesn't - then suspect a bad op-amp, and replace Z23.

If both Z23 pin 1 and pin 3 look 'bad', then check the sample and hold caps (C28/C36). If they look OK, then we'd need to look farther back towards the muxes (Z19 and/or Z20).

If it was found that Z21 pin 7 (Pitch A) and Z23 pin 1 (Pitch B) were both OK (and if the TP2 signal is bad, but the TP1 signal is good), then we are looking at some of the electronics between the Pitch CV in, and the TP2 output.

Excluding discrete components, that generally equates to part of Z1 (CA3086), part of Z4 (4558) Z3 (4151) and part of Z6 (LM393).

The good news is that all of these parts are cheap and easy to get. If worse came to worse and you had to swap out Z1, Z3, Z4 and Z5, you'd probably be looking at less than $5.

As a starting point I would lok at the output of the charge pump on both channels (Z2, pin 3 for voice A and Z3, pin 3 for voice B). If they look about the same, then suspect Z6. If they look different, then the problem could reseide in Z4, Z5 or Z3. (or a discrete component, etc.).

Good luck with this card. Let us know how you make out.

In another voice card both the output from TP 1 and 2 are too low, not an octave higher as espected from the procedure like working voice cards.

A good starting point would be to trace back the control voltages here too (just as in the case above). If they control voltages are too high/too low, that would explain the frequency error at TP 1 and TP 2. You can then look to see if that's due to the actualy Sample and Hold circuitry or something else.

For these investigations, be sure to take a peek at the Dual Channel Board Schematic, pages 6-10 and 6-11 from the Service Manual.

Sandro Traversi [21010217] · Thu, 16 Oct 2003 15:14:06 +0200

I totally agree with Dave (a master!).

Maybe I'd look first to Z19 and Z20 and swap them. I've found broken so many ...

See Fixing Voice Card update (December 2003)

Attack Times

Chris Ryan [21030691] · Mon, 20 Oct 2003 08:01:02 -0700

I noticed this message posted to Analogue Heaven. Anyone know the attack time for the Chroma?

Date: Sun, 12 Oct 2003 18:47:36 -0400
From: Rob Belcher
Subject: Re: Envelope speeds.....

Hello people,

Im interested in the Attack times of various synths... Im aware that the envelopes on hybrid polysynths arent as snappy as say the Prophet5 or Jupiter8 which are fast, quoted at 1MS.

The Prophet5 and JP8 obviously use analogue envelopes, unlike the Prophet 600 or T8 for example, which are digitally generated... i can see how early digital processors may not be up to the job of scanning the keyboard, the panel AND generating a quick envelope..

However what im particularly interested in is some of the hybrid polysynths that can do the above and STILL generate a quick envelope, the Synthex is quoted as having a 1MS attack time.... What slightly puzzles me is why could Elka manage it on the Synthex and SCI couldnt manage it on the T8, which must have much more processing power in it than the Synthex?

Id be interested to know if anybody knows the attack times for the below synths, it would be most interesting.

  • Elka Synthex
  • Rhodes Chroma
  • SCI Prophet T8
  • SCI Prophet 600
  • Oberheim Matrix 12
  • Oberheim OB8
  • Oberheim Matrix 1000
  • Roland Juno 106

These are not all the analogue synths, with digitally generated envelopes, however they are particularly the ones im interested in. If nobody has specific information, can anybody give an impression of how quick the attacks are for the various synths, perhaps put them in some sort of order, snappiest first.

I find it quite interesting as i like snappy sounding envelopes, and it would be good to have some sort of information about this on AH.

Regards,
Rob.

David Clarke [21030085++] · Thu, 23 Oct 2003 22:16:44 -0400

Empirically the Chroma doesn't seem as 'snappy' as something like the Odyssey or the Prophet-5.

I could certainly do a few measurements, but does anyone know what the real 'figure of merit' is? That is to say, what is it that would really define the 'snappiness' of the attack - would it be the initial width of the attack pulse, or something else?

Christopher Now · Fri, 24 Oct 2003 15:03:46 -0400

David -

One possible interpretation of snappiness would be the response of the instrument to a note depression (the time between note down and sound at the output). Using a dual-trace scope set to measure the output of the Chroma tirggered relative to either incoming MIDI or from the keyboard scanner CPU. The synth in question would use a raw waveform with all amp and filter attacks set to zero.

I've never measured the Chroma, but I did assist Keyboard magazine several years ago when they were gathering data on how 'fast' samplers were, relative to a note-on/key down. We used the incoming MIDI signal as the trigger to measure the time between receiving a note on and actual sound at the output. Since the Chroma uses no D-A, it *should* be able to have 1 millisecond or less of a response time.

The Chroma has a very fast keyboard and I've always felt that it was pretty 'snappy' for my needs.

Don Tillman · 25 Oct 2003 00:07:28 -0700

One possible interpretation of snappiness would be the response of the instrument to a note depression (the time between note down and sound at the output).

Sure. Plus some limit on the risetime.

I've never measured the Chroma, but I did assist Keyboard magazine several years ago when they were gathering data on how 'fast' samplers were, relative to a note-on/key down. We used the incoming MIDI signal as the trigger to measure the time between receiving a note on and actual sound at the output.

I'm really curious... what were the results?

Also, don't all samplers available on the market encode their keyboard into MIDI and then pipe that MIDI line into the microprocessor? Then a proper measurement should also include the time it takes the keyboard to queue up and spew out the MIDI commands.

The Chroma has a very fast keyboard and I've always felt that it was pretty 'snappy' for my needs.

Just from the feel, I would guess it's about 1/4 the response time of a MemoryMoog.

Rich Hilleman [21030351+] · Sat, 25 Oct 2003 11:31:32 -0400

I agree with Don.

When I have measured Guitar Synths, response time is what they are obsessed with. I would capture the Guitar's signal on one audio track in PT, then the Generated Synth Signal with the best EG settings and measure the latency in Samples. It worked pretty well. I have tried dual trace oscopes, but this was easier. You could probably use the Keyboard Control Voltage as an audio level and do the same thing.

Josue Arias [21010162] · Sat, 25 Oct 2003 15:58:41 +0200

HI People,

Now I remember a Keyboard mag from about 10 years ago titled: "Timing is everything" It have an extensive article and a table of near all synths and the MIDI delay response. Does anyone have this article? Could be great post it in the www.

Christopher Now · Sat, 25 Oct 2003 16:24:56 EDT

I'm really curious... what were the results?

If I recall correctly, the Akai samplers had the least delay and the least jitter. This was in the era of the S900/S950, Emax, Prophet 2000, Emulator II, Korg DSS-1/DSM-1, Mirage, etc. I do know that there were NO offsets for the samplers' internals (speed of amps & filters) taken into consideration for the results. The S1000 and E-III may have been included in that batch, but it's been a long time since I thought about it. I should look for the article - I don't even remember what the exact year it was, but I doubt it was later than 1991-1992. Keyboard may have done a later article that compared the response time of synthesizers.

The motivation came from finding out what unit was best for use as a percussion playback module when triggered from a MIDI pad. There were a lot of public claims at the time that MIDI cables caused delay and Keyboard wanted to address the real issues of what people were hearing or feeling and calling "MIDI delay".

Also, don't all samplers available on the market encode their keyboard into MIDI and then pipe that MIDI line into the microprocessor? Then a proper measurement should also include the time it takes the keyboard to queue up and spew out the MIDI commands.

Yes, that was certainly true of the '80's/early '90's samplers, but to even the field, the testing setup was done with every device a slave of MIDI because one of the samplers didn't convert the Keyboard to MIDI before it reached the processor (maybe the Kurzweil ? my memories may be suspect on specifics) I work for a company that developed sample editing software for these samplers, and we had a studio with all of the 'current' models of the time (1985-1991). I noticed the other day that one of those samplers was being used to prop open a closet door!

I'll see if I can locate this article (the next time I'm in my attic) :-)

David Clarke [21030085++] · Mon, 27 Oct 2003 21:32:22 -0500

I just did a couple measurements.

There looks to be 3 - 20mS of 'keyboard' lag. That means there can be as little as 3mS and as much as 20mS delay after a key is struck until audio output appears. One big factor concerns the keyboard algorithm chosen (for instance, hitting a single key in Polyphonic mode is approx 3mS, but using the same program but with a ALL CHAN poly keyboard algorithm selected resulted in a 20ms delay to the first audio out.)

I also looked at the min/max Attack/Decay/Release rates.

With "Attack" set at minimum, the AD(S)R attack ramp time is approx 4.7mS (i.e., 4.7mS from the start of the attack pulse to reaching the full attack control voltage).

It is interesting to note that the service manual points out that a capacitor was specifically added to slow the VCA attack response of the voice cards ("...the 0.1uf capacitor on this input slows the control voltage changes down just enough to make a sharp attack or release sound like a soft pop and not a sharp tick...").

The approx. 5mS min. value is almost exactly what you'd expect just from that value of capacitor alone.

Out of curiosity I did a measurement with this cap. removed. As would be expected, in this case (with the panel parameter set to 'instant') the VCA control voltage is immeasurably sharp (no lag at all).

I saved some plots from the 'scope, as well as tabulated the attack times, etc. I'll keep these on hand and when I get a few moments I'll try to jot down a few notes to rationalize the values (e.g., delay x is because of 1024uS program loop so many times, etc.)

Erik Vellinga [21010286] · Tue, 28 Oct 2003 17:26:10 +0100

As an addition to this; I am very much interested in the frequency spectrum of the Chroma. I can remember that some years ago I measured the output with a scope and found that the spectrum rapidly decayed above 7KHz. I used the self-oscillation of the filter and swept it into the high range with the filter tuning. Funny that one of the reasons the Chroma is famous is it's crispy sound... Right now I do not have a spectrum analyzer available but I may have a chance to get one at my work later this week. Can anybody confirm this? Maybe there is a defect in my Chroma's audiomixer??

Greetings from Holland,

Erik

Åke Danielson [21010068] · Wed, 29 Oct 2003 08:59:00 +0100

Do I understand this rightly? Can you make the Chroma more percussive by removing this capacitor. I always found the Chroma's slow attack time to be its biggest drawback. I capacitor removal is the solution, I'll get rid of them all!!! Where is it located?

David Clarke [21030085++] · Wed, 29 Oct 2003 22:47:16 -0500

Do I understand this rightly? Can you make the Chroma more percussive by removing this capacitor.

Yes, possibly.

The biggest change will be seen when the VCA Attack, Decay and/or Release are set to their minimum settings (i.e., the "instant" setting). I would hope/expect that this will allow a much faster initial response.

...I capacitor removal is the solution, I'll get rid of them all!!! Where is it located?

There is one capacitor per channel, meaning there are two capacitors per voice board. The capacitors in question are C43 and C44 (0.1uF), connected to pin 5 and pin 10 of Z16 (respectively).

The location of these components are shown in the following picture (the blue/green capacitors pointed to by the yellow arrows).

If you did wish to experiment with a modified board I would recommend only making the change on one card, and removing all other voice cards (so only 1 voice card - a modified voice card - is present). In that way you'll be able to determine if the changes are audible.

If you do decide to try this, I'm sure the whole group would be iterested to hear what you find.

Rich Hilleman [21030351+] · Thu, 30 Oct 2003 11:47:50 -0500

Uhhh.....

I think I would consider this a little before making this call.

I think that there are a number of "Unusual" design choices on the Chroma. The VCO section is not implemented the same way that I would have expected, and clearly they made the filters sound dramatically different than any other synth using the same components.

I have three Chroma's and a Memorymoog, OB8, Prophet 5, Jupiter 8 and Waldorf Wave, all with Curtis chips. I love how they sound different and use them for different things.

I think this differences are what make the Chroma unique. I suspect you will lose some of that character if you go too far down this road.

Don't lose what you love while you pursue more. On the other hand, I have this Hot Roded Arp Avatar, so who am I to talk.

Rich Hilleman [21030351+] · Thu, 30 Oct 2003 22:32:23 -0500

Registry

I have two keyboards and an expander. I thought you had them all, but I will look. It is possible that you don't have one of the keyboards. It has been at my Tech (CAE Sound) for a year while he uses it to get his Chroma restored. Pesky keyboards.....

David Clarke [21030085++] · Thu, 30 Oct 2003 22:20:48 -0500

Do we have all three in the registry? (I can see two there [21030351 and 21030800] - are we missing one?)

See Attack Times (redux) in the December 2003 message archive.

$300 Chroma

David Clarke [21030085++] · Thu, 23 Oct 2003 22:20:36 -0400

Folks - I was poking around the Chroma site today, and noticed that one of the Chromas on the "ads" page is only $300 - and that's with working sound cards and pedals!?

For $300 it's a hard deal to pass up - but I need another Chroma like a hole in the head.

Does anyone know of a good home for it (so I won't be so tempted anymore :->)

Csaba Zvekan [21030495] · Fri, 24 Oct 2003 10:43:57 +0200

I might be interested, since I could get mine to work.

Fwd: It's All Done!!!

Chris Ryan [21030691] · Sun, 26 Oct 2003 19:04:42 -0800

Some good news from Paul ...

From: "Paul Hackett-Evans" [21010094]
Date: Mon, 27 Oct 2003 02:32:48 -0000
To: Chris Ryan
Subject: It's All Done!!!

Hi Chris, Great news!

A few days ago, the UK Chroma engineer, Roy Paynter, phoned me to invite me to his home, where he has his workshop, to see the results of his work on my Chroma, 21010094. Roy's had my Chroma for the best part of 2 years, not because there was any problem with working on it, but because he wanted to source the best components and it took time to get them.

Roy was so happy with the way my Chroma was shaping up, that he asked if he could hold onto it for some time to use as a benchmark for the other Chromas he's had in this year, notably 21010250. Peter Riani's Chroma was being restored after it was dismantled by students at the Newcastle School of Music, UK, and the parts rescued by Pete from a rubbish skip. Roy's had the power supply, the I/O and computer boards from Pete's Chroma and he was using mine as a template for the restoration of Pete's boards. Now he's finished them, ready for shipping back to Pete, so my Chroma's ready to come back home, too.

Well, today, my 6-year-old son and I visited Roy. He proudly took us into his workshop, where my Chroma was on a keyboard stand, plugged into an amp, ready and waiting.

Roy switched it on.

It's been such a long time since I saw my Chroma working, that I was thrilled to see the LEDs all blinking together in sync as the Chroma booted up. It was lovely!! I savoured the moment, no error messages in the display, no noise, no hiss, no hum, just quiet. I was already falling back in love.

"Go on", said Roy, "try it."

I touched a key. Struck it. Played a chord. The background noise was so quiet that I wasn't expecting the massive, awesome, colourful, living sound that blasted out of that amp. Wow!! If I needed to be reminded why I always thought this was the best synth to come out of America, that moment brought it all back. My son, Alex, who'd never heard a Chroma before, was bowled over. The sound is full, vibrant, fresh and alive, with all the organic qualities that made the Chroma stand out for me when I first heard one back in the 1980s. The bass went solidly down into subsonics, and the highs were clear as a bell. Roy's done a magnificent job on the electronics, and my work on the physical and mechanical restoration of the cabinet and the keyboard action and its contacts are satisfying, too (remember I hadn't seen it for 2 years!).

Let's just say it was a long time before we powered down the Chroma at Roy's. The man's a genius, and such a nice gent, too.

152 miles further on, and several hours later, the Chroma's back at home, resting in its flightcase next to me as I type. I'll have to rearrange some furniture to give it room to be set up and played, but Roy said that after his and my labour of love to get the Chroma looking as good as it does and sounding better than it ever has, he feels it deserves to be played and enjoyed. And it will be!

What a happy day!

Paul

chroma midi F.S.

Roger John Lesinski Jr · Wed, 29 Oct 2003 23:52:12 -0500 (EST)

Not mine

Syntech/Chroma Cult MIDI for Rhodes Chroma Item number: 2569130123 ebay