A Search for the way to restore the original preset software to “bricked” units of the limited edition legendary DSP 4500

To be more precise, the 4500 FW used flash memory chips on which the software for the effects was stored. What’s needed is a working, programmed flash memory chip from which the 4500’s software can be copied. The software can then be installed (‘flashed’) on other 4500s whose flash chips were inadvertently cleared. Eventide discarded all its copies of its own old software when they superseded the 4500 with new models. If anyone has a working 4500, please please please respond so that we can resurrect this valuable, irreplaceable gem.

For those wondering how the chip got cleared, it seems that in the old days Eventide had a key sequence that could clear the chip so that they could re-program it if it wasn’t working properly… back when they still had the code to re-program it.

Wouldn’t installing already flashed chips/copies do the trick? I’ve seen these around more than once. Wouldn’t that make this any easy fix, since people are already providing these? Just a thought…

As far as I remember, the 4000 series still used replaceable memory. Eventide may not have it available, but its out there if it’s just the programs you are missing. Forgive me if I am not understanding the problem correctly…

Here’s the DSP4000 ROM, Modules, and PROGRAMS (I haven’t looked inside a 4500 lately…you’ll have to look)
http://studiorepair.com/gallery/Eventide/DSP4000/slides/Eventide_DSP4000_ROMs__STUDIOREPAIR_08101301_1009298870.html

Who is the expert who wants to reverse engineer the memory?  If there is that big of a demand, why not start a go fund me or something?  I am checking out a used 4500 this week, but its extremely low price makes me think it might already be DOA.

Please let me know if you pass on the used 4500, as it might be possible it has the missing ingredients to restore the bricked unit sitting in my workshop. And thanks for the suggestions.

I recently saw an old Eventide technical note document that said “In the DSP 4000 series, where some features are available on plug-in cards, an EPROM OS upgrade was furnished with the purchase of the card.” Is it possible that such a card might be able to bring the bricked 4500 back to life? And where would one find such a card? Any help appreciated! thanks

Through a series of misunderstandings I now have the unit(at no cost) but I do believe it is a brick.  I told the owner of he wants it back, he can either pay me for shipping or come get it.  ROSINCORE, If you’d like to discuss it further, please contact me in private.  Is there a way to do that on here?  I would be willing to buy it off the guy(for a  further reduced price) if you think it could help the community at large, on the hopes that mine may be fixed as well.

Hi Epiphoneedge — I would like to discuss this with you. Does the unit turn on at all? There doesn’t seem to be a private message conduit on this forum, so why not just email me at my work address, neocortex2@verizon.net. Looking forward to hearing from you.

fwiw, still have a working GTR4000 if that helps. great unit!

Allright! This topic has been coming up in different places quite many times lately and thus requires some explanation.

I suspect there’s one or two bricked units around, DSP4500. These are floating and changing hands quickly at cheap prices because of this problem.

Jake, the thread starter has it right. The 4500 was the very first Eventide running system, modules and presets on flash memory so everything there is on NON

removable flash chips, not removable EPROMs. So… don’t even hope to copy EPROMs like it used to be the solution for many other units.

The 4500 on flash was the beginning of this different memory technology that very quickly brought to the next step, the DSP7000/7600/Orville marvels.

OS updates were now fast and cheap as basically it became a matter of downloading a self expanding application that would update your unit.

The problem with the 4500 is that it never had any OS update as it was a very short life product before jumping on the 7000/Orville series, a much more

powerful platform. A 7000 was 4x more powerful than a 4500. Orville was 8x. So no more reason to keep the4500 alive.

This leads to the fact that there was never an  application to download and update a 4500, nor to fix it. The code might be in the Eventide hands or simply lost for

good. Nobody there will do anything for a bricked 4500. That’s sure like night and day!

Solutions:

-what Jake says… but it would take some extremely knowledgeable person in even trying to do what he describes in his first post above. I’d just say “forget it”!

-a voluntary with a working DSP4500 who has the immense courtesy and abundance of time to download each single preset off the machine in MIDI SysEx files.

IF possible, these would need to be loaded ONE BY ONE on the bricked bastard and stored to a plugged in PCMCIA RAM Card. You’d need 2 or 3 4MB of  

these hard to find cards, but they’re still around. The machine can take up to 4 MB, not bigger. You’ll live by swapping cards to load a desired preset.

Or you simply keep the files on your computer and use a MIDI SysEx app (Sysex Librarian for Mac – Midiox for Windows) to quickly send any preset to the

4500, as its own memory disc. BEST SOLUTION to me!

-MAYBE and I say MAYBE AGAIN there’s a possibility Eventide still has the factory presets in sigfile format. These would need an installation of the old Vsig

program, previous to the later V2.x, and you could upload single presets to the machine but will also need the PCMCIA cards to store them. The internal RAM is

just very small. But Vsig 1,xx would only run on very old Windows systems, like up to XP. And running Vsig, the old one, isn’t for everybody out there.

Other ideas….

Any DSP/GTR4000 EPROM won’t help. Maybe their presets in MIDI SysEx format could help.

The ULTRASHIFTER V2.300 card has some modules and presets updates/upgrades in it, BUT it was never meant to be a “fix your DSP4500” tool. ‘m not sure it

would do any good at all. It was never tested as this issue solution and is very, very hard to find. It added the Ultrashifter presets, the MIDIclock presets, the

Dither presets, a set of  modules.

Any other DSP/GTR4000 card won’t do anything at all. No systems are on board.No flash capabilities.

design file archeology was always hit or miss at my company, all the original source versions, programs that’ve changed, formats updated/adopted/abandoned, operating systems, and design tools across a couple of decades are too many to articulate. but yeah, flash presents *issues* that EPROM didn’t. as in lots of.

So,

I finally got the unit home to my humble studio, fired it up and had it 2 errors.  I forgot to get the one, something to do with ROM i think, and it displayed the other time i first turned it on.  but the one said “ERROR – SUGGEST POWER DOWN, host timeout”.   So it fireup, and i load up the banks, and choose delays, and choose a simple stereo delay(number 2 perhaps).   I set it up with 2 aux feeds from my desk, with a simple demo from my last band.  AUX 1 into #1, Aux 2 into #2.   The output is routed back to 2 empty channels.  I see signal is being passed to the unit, as the front input meters light up in time with the song.   But no audio is being output.  I check the return cables via cable tester, both pass.   So now im RTFM(not my first time mind you, but first time with the unit in front of me), and i see there is both a global output and a preset output.  So the global is set to 0db(max), as is the preset output.  I turn up the preset slowly, and nothing.  I tried this for a few presets, in different banks.  Unfortunately im out of time for the night, i will continue this on Friday.  Fortunately, my job gives me ample time to RTFM.

So is this how a bricked unit looks/acts?    I can see the banks/presets, there are parameters that can be adjusted, the input meters show incoming audio, but nothing from the outputs….unless i change a preset and then it gets LOUD for a split second.  I will also be trying to get a sysex file, and see if it will load via MIDI.   IF it does,  i will probably keep it, and start begging for and buying presets.

epiphoneedge wrote:

So, I finally got the unit home to my humble studio, fired it up and had it 2 errors. I forgot to get the one, something to do with ROM i think, and it displayed the other time i first turned it on. but the one said “ERROR – SUGGEST POWER DOWN, host timeout”. So it fireup, and i load up the banks, and choose delays, and choose a simple stereo delay(number 2 perhaps). I set it up with 2 aux feeds from my desk, with a simple demo from my last band. AUX 1 into #1, Aux 2 into #2. The output is routed back to 2 empty channels. I see signal is being passed to the unit, as the front input meters light up in time with the song. But no audio is being output. I check the return cables via cable tester, both pass. So now im RTFM(not my first time mind you, but first time with the unit in front of me), and i see there is both a global output and a preset output. So the global is set to 0db(max), as is the preset output. I turn up the preset slowly, and nothing. I tried this for a few presets, in different banks. Unfortunately im out of time for the night, i will continue this on Friday. Fortunately, my job gives me ample time to RTFM. So is this how a bricked unit looks/acts? I can see the banks/presets, there are parameters that can be adjusted, the input meters show incoming audio, but nothing from the outputs….unless i change a preset and then it gets LOUD for a split second. I will also be trying to get a sysex file, and see if it will load via MIDI. IF it does, i will probably keep it, and start begging for and buying presets.

I haven’t personally seen a 4500 with factory presets erased but your description looks odd to me. If presets were erased you shouldn’t see their names, nor banks. The factory test that caused the problem would have completely erased the presets flash. So I tend to think your unit has other issues. One of the problems could be the internal battery. 4500 is about 25 yrs old… that might be something you could do.

I’d also try a CLEAR SETUP and load the THRU preset, checking if audio passes thru I/Os, without hitting the DSPs.

Whatever you do, DO NOT, I repeat, DO NOT run the self-tests described in Technical Note #47. (These tests are initiated by holding down the 5 key during startup). As the note says, many of these WILL ERASE FACTORY PRESETS and will leave you with a useless harmonizer for which there seems to be no “restore factory presets” software/algorithm/what-have-you. I learned this the hard way. What insanity prompted Eventide to code these “self-tests”– more like ‘self-destructs’ — in the first place perplexes me, unless I’ve totally misunderstood this… I mean, how did does one put preset code on the flash chips in the first place? There’s gotta be someone out there who knows the answer. I just wish I could find him or her.

Rosincore wrote:

Whatever you do, DO NOT, I repeat, DO NOT run the self-tests described in Technical Note #47. (These tests are initiated by holding down the 5 key during startup). As the note says, many of these WILL ERASE FACTORY PRESETS and will leave you with a useless harmonizer for which there seems to be no “restore factory presets” software/algorithm/what-have-you. I learned this the hard way. What insanity prompted Eventide to code these “self-tests”– more like ‘self-destructs’ — in the first place perplexes me, unless I’ve totally misunderstood this… I mean, how did does one put preset code on the flash chips in the first place? There’s gotta be someone out there who knows the answer. I just wish I could find him or her.

Those tests are actually necessary. The ones erasing presets DO warn the user… so one has to be clever enough to avoid doing what’s not to be done.

The chips were likely flashed via COM port after being installed. The 4500 was a very short and limited production. things on flash really started out with the

later generation.

Gentlemen,

So i got quite a bit of time in with the manual at work just reading and re-reading, and I started to suspect I misunderstood the BYPASS/MUTE/KILL functions.  So back home, I fire it up, make a few changes and….there it is!   Now other adjustments needed to be made(levels, etc.), but yes I have a 100% WET signal.   I’m not sure its optimized for my setup, as the settings are the previous owners(minus a few things i changed),  but i have the manual, so Ill figure it out.  And i wont do a factory reset just yet.   If I understand things correctly, I should be able to copy the factory preset sysex files, and then email those to whoever wants them.  That being said this isn’t going to happen tonight or tomorrow.  I will probably release them as BANKS.  I guess it depends on how large the files are, compressing them, etc.  I’m not the most tech savvy, but i should be able to get this done.  I have loaded files onto a unit via sysex, i imagine it shouldn’t be too hard to copy them off of one.  As always, if anyone has any advice or ideas, feel free to say.

Sorry, didn’t mean to offend by my rant… I’m just frustrated that I’m not more clever, and I wanted to help by making sure that Epiphoneedge didn’t make the same dumb mistake I did. But here’s a thought: if the chips were flashed after installation via a COM port, could that same port be used to re-flash them after erasure? And (sorry to be really really not clever) what’s a COM port? I don’t see any ports on the back of the unit other than XLR jacks, MIDI jacks, and S/PDIF jacks.

Rosincore wrote:

Sorry, didn’t mean to offend by my rant… I’m just frustrated that I’m not more clever, and I wanted to help by making sure that Epiphoneedge didn’t make the same dumb mistake I did. But here’s a thought: if the chips were flashed after installation via a COM port, could that same port be used to re-flash them after erasure? And (sorry to be really really not clever) what’s a COM port? I don’t see any ports on the back of the unit other than XLR jacks, MIDI jacks, and S/PDIF jacks.

No problem, Rosincore. I understand the frustration!

The COM port is an 8 pin RS-422 connector. This serial port was implemented in ALL later units (DSP7000/7500/Orville/H7600/H8000) to connect the Eventide to a Windows PC running Vsigfile Graphic Editor, the tool used to create algorithms and to flash the internal chips with OS updates available in time.. These ports were not always installed on the 4000 series and often the machine came with an expansion slot (or two) for them. My 4000 had them. The 4500 had a single remote panel where the COM port could be installed.

The fact is that even when present they were never implemented on the 4000 series, so all Vsig programming was possible via the MIDI IN and OUT ports.

Eventide may have used some kind of tool with sockets for the flash chips, connected to a PC via serial cable/COM ports to flash a bunch of them at the same time. They didn’t flash ’em in the machine. My assumption is incorrect.

SEVERAL MONTHS LATER

Ok, sorry, life got in the way, and music took a back seat.  So I finally have some time to get things going, and i ran into an issue or two where it wouldn’t load anything, but i knew the presets are there.   So I unseated the (i believe) the ROM(it says BQ4013YMA-120), cleaned that out, and a few other socketed chips, and….well, it works again.   So i have downloaded (DUMP) 2 files off of my 4500 just to make sure I’m doing this right.   They are both under 10K in size.  If you have a bricked unit, or can verify the integrity of the filess, please send me a good email, i will email these to you.  You will then have to load them onto your unit. I used USB to midi cables and MIDI-OX.  it might take you a few YouTube videos to learn how, but it isn’t hard.    Of course, thought and comments are always appreciated.

@epiphoneedge Hi, so you solved the problem isn’t it? I’m in a similar situation with a Dsp 4000, it sometimes (few) works, other it stops with the ERROR: Hc, power down and so on. You unseated and replaced the backup battery? Because afaik that bq4013 chip is a battery, not a Rom with preset, but I can be wrong…

Thanks

Lucio