But there is nothing availible today like the H3000. No company with the expertise, resources and dedication of 1980's Eventide is currently making digital units with R2R DACS feed by clocking allowing for variable samplerates and featuring analog feedback paths between DA and AD.

You can either get this glorious sound at a steep used price or buy a plugin (running in a DAW or a dedicated box) that sounds bland and pedestrian in comparison.

Well the H3000 was not the first Eventide hardware to use variable sample rates on ADC and DAC combined with analog feedback. AND it’s not the last Eventide product to work this way. That’s exactly why our Rose pedal sounds like it does.

Cool, I didn't know that. But it's not a studio multieffect unit, so not in the same market.

Its was a great box for teh $800 I paid but not worth current prices IMO..

I agree Rose is not a studio multieffect unit. Didn't suggest that however with long memory, delay multiplying, reversed delay and analog feedback plus line input it is something special.

Ok! I've had enough of that sluggish preset load response!!!

Some time ago I have built this table which shows which IC contains which algorithm and preset. What was left to do was to see which presets are sluggish and mark them in the "black" column. And here is the result: https://drive.google.com/file/d/15dK...ew?usp=sharing

And it immediately dawned it on me! We have duplicates! Take a look at the table. Many of the algorithms are present twice! So each time we load the program, the machine loads the algorithm two times. Right?

So I took out my EPROM burner and EPROM eraser and decided to put an end to this. Started the HEX editor and spent 2 days staring at the binary data - a hex code. I know there are duplicates in the code, but you can not just copy, paste or shift data any way you want. This binary data. These are direct instructions to the DSP. One character too few or too much and the rest of the code is dead.

Very frustrating. I had to learn the patterns: How does a DSP approach the algorithm, how does it differentiate it from a preset, what is in between two algorithms or presets, how is the "empty" space filled in the binary file, how does DSP know where algorithm or preset starts and where it ends. All that without source code is very difficult to figure out. Sow now you see why it takes 2 days, rather than 2 seconds (to select and cut some duplicate data).



And slowly I cleaned up this mess of doubles / duplicates. Also used the opportunity to move the presets from one IC to the other as I've noticed some empty space. By that time I learned how to add a preset into the binary code without ruining the rest of the binary data. So I eventually scaled down the B-DFX upgrade from 5 chips down to 4.

Fortunately I have the UV EPROM eraser, and EPROM reader/writer so I can make as many test copies as I want. And after a couple of more hours I had what is cleaned up, fully functional, B-DFX upgrade that now takes just 4 ICs (instead of 5). I was certain that I have made it. But I was wrong. The unit is still sluggish. Facepalm!

So it must be something else. We are missing something. That Broadcast upgrade is suspicious. It does not contain Timsqueeze algorithm. So it must be some modified or special broadcast upgrade. I give up.

*Regarding the modified B-DFX upgrade, sorry I can not provide the binary data. Too many hours spent compiling and shifting this data to be given away for free, yet I can't sell something that does not belong to me neither. So yeah. Sorry about that. Build your own.

Oh boy!
Don.... what about the chips?
Those might be the issue....

Rose looks and sounds really nice. Definitely want one. I hope Eventide continues to work the elements of the H3000 that make it sound so special into future releases, in lieu of the impossible task of reissuing h3k 1:1.

I used fastest possible, I think they are rated 10ns.

https://reverb.com/item/53852835-eve...-just-serviced

Just noticed this not-too-horribly priced DSP4000 on Reverb if anyone is feeling frisky...$1599 US

Mhhh...
not sure about that display...
expensive....
hopefully it runs on OS V2.158 or.... leave it there....

(Not a critique towards you, but the sellers.) Sorry but this has nothing to do with H3000. There is nothing "H3000-like" in the DSP4000. These DSP4000 sales come from the folks: "it's old so it must be as cool as H3000", except that it isn't. Don't let these tricksters fool you.

When this thing came out it was all bells and whistles about how new and advanced it is compared to the old H3000. DSP4000 and later are all based with a built-in patch editors to let you create your own effects algorithms, and with each iteration they packed more and more DSP power. But the sound is essentially the same. So you can grab Orville which is essentially the same thing except it has quadruple power of this thing if not more.

The only non H3000 that sounds like the real H3000 is H3500 and that's where that line stops.

Thanks Don, to be fair I've never used one and it simply looked like my H3000B so, live and learn.

DSP4xxx to H8000 ALL sound really amazing! Different from the H3000 but still way top notch and even much better than the H3000 for several tasks like mastering. But yes... that price is very close to an Orville that would be 8 times more powerful than a 4000, reach 96KHz sampling rate, run two algorithms at the same time, open to full algorithm design.
And of course it also runs like 32 to 36 H3000s in a single box....

Orvilles seem to be shockingly rare on Reverb - only four sold when you search their sold listings (in the US) for some reason. Surely more than that have gone up for sale.

Orville is a keeper!

Had one like, the H8000 better...

As I said before..it has nothing to do with the eproms. I had an original h3500DFX-B factory machine...it is just as slow as a blackface H3000 with the kit in it. It has nothing to do with RAM IC next to the eproms...again it is the same in both machines and the same speed.

But can you tell me what is the actual problem in practical use?? I can't detect anything but scrolling through some of the presets slowing a bit which could be down to something like display ram or scratchpad.

My original H3500 DFX came with factory eproms that had no sluggish response. Stupidly I threw them away after buying the B-DFX upgrade from ebay. I should point out my H3500 is an early unit, meaning it had less presets than later DFX, hence why I decided for an upgrade.

In this table:
https://drive.google.com/file/d/15dK...QZk-bfMb3/view

It is listed as DFX*, with the asterisk indicating the early version, as opposed to later DFX series, presets are shown in cyan color. It's missing a tone of them.

Ok I see, but I also want to know exactly what the problem is you are seeing on your machine...sluggish in what way exactly?

DP4+ hum: I had this too. On top of the transformer in my unit there was a metal bracket, thin metal, screwed into the transformer on two sides, and also screwed into the top panel of the DP4+ case. I removed it, having determined that that tiny bracket can have no heatsink function. All humming/buzzing disappeared.

I recently realized that it might have some grounding function, though it seems unlikely. Will attach a simple cable between two screws, one attached to the transformer and the other to the top of the case. Will report back in case this electrocutes me a little…

I am at preset 255 I want to fast forward 15 presets. I move the main dial, the response after certain number ie 259, suddenly drops that I can no longer use the dial properly as it will jump over the presets, skipping some, slowing down, no longer moving linear way. The thing has 1000 presets it's not that easy to memorize them and just type in the preset, but you know the general location of certain presets, hence you use the dial.

Now... when someone spends his time wiping the dust off the H3000 or admiring typefaces of the front panel, this all probably doesn't matter, but when you work in a busy studio environment under a bit of pressure, it becomes a bit irritating.

Most likely the bracket is there for additional stability for the transformer when the unit is being shipped/moved around. A grounding path could also be a secondary function of it.

Over the years I've seen way too many synths/effects units get damaged during rough shipping because a transformer broke loose and bounced around inside.