PitchFactor 3.0.0beta[16] & FactorLib Testing

[brock]

I happened to stumble upon the newest beta version and new (?) librarian version.  Here's a not-so-quick summary of what I've experienced so far:

I wasn't certain of my updater version, so i downloaded the latest and ran that version.  For some reason, I had to re-input my password to avoid repeated download failure.  From there, it was "grab the popcorn and watch the show".  Kind of interesting to compare the software program progress feedback to the display and LED indications on the box.  I'd guess about three minutes total from login to "DONE !"

I intentionally left some custom presets in place, to see what might happen during new beta migration.   The presets remained intact, as did the preset naming from the 3.0.0beta[0].  So did my MIDI settings … at first.  After the next "cold boot" (the first one post-updating), all of my MIDI channels, CC assignments, etc. seemed to revert to defaults.  The CC assignments were't what I remember as defaults from the UM (PDL 99?),  But, hey.  I changed them back to "my" MIDI settings, and the parameters held through another few reboots.  Easy fix.

The first thing that I tested was BND as a MIDI RCV parameter.  I didn't think that it worked (at first).  In fact, I first thought that all MIDI assigning was "broken".  I was getting CC's or BND "learned", but the display wasn't reflecting that.  The same result happened if I dialed in the assignments.  BUT … after moving the targeted knob physically, the display started picking up my MIDI parameter changes.

MIDI Pitchbend control is fixed!

The PitchFactor had been orphaned from my normal setup for updating, so I wasn't getting audible results (just USB MIDI, a small MIDI keyboard, and my DAW during the update).  I'll give it the full treatment after this initial  "connected" testing.

That fine-tune control over ANY parameter works flawlessly (nice touch, ET).

The Level Meter is a cool addition.  Two switches pressed at the same time, and the Meter only stays active until you release them, but I think that could work out well (hands-free) for gain-staging your input levels.

It appears that the Synthonizer waveforms are now distributed a little better than before.  The OR1 and OR2 waveforms aren't 'squeezed' at the end of the knob's travel. 

I've done very preliminary testing of the FactorLib / beta[16] interaction.  I thought that there might be preset naming in software, but I'm not getting that in test scenarios.  Perhaps it's user-error at this point.  Under WiinXP, there are some issues with window resizing and Windows conventions.  Maximizing, yes, but resizing is flaky.  But it DID upload accurately from my Pitchfactor (ALL or single presets), and even loaded up (and sent) single pressets that I had saved using Bome's SendSX, then transferred to the PF.  So that's worth the 'price of admission' alone.

I'll be moving the PitchFactor back "home" now,  so I'll flesh out this thread as the real testing gets underway.  I'd welcome anyone to add their own experiences here.

[timothyhill]

Thanks for your detailed post! I'm running Win7 and didn't have any of the Windows problems you reported. That may be related to differences in WinXP and Win7.

I was able to rename presets in FactorLib by double-clicking on the preset name. Again, that's under Win7, so I'm not sure what WinXP will do.

The Edit Console seems to be working, too, which is great!

As you, I've had little time to do more than preliminary testing. I'll add more as I go and I'll be looking forward to your results, too!

[daggodag]

I also dowloaded V 3.0.0beta[16] last evening.   The update went OK, but a few presets changed quite dramatically.  Three of these were user presets and it appears that there was a change to feedback settings that made everything squeal badly.  I was able to manually change the settings back and things were fine.

Presets 6.1 (Anthem), 15.1 (910 Chorus) and 15.2 (chorus room) also had problems.  The pitch settings changed  6.1 and the feedback problem occurred in 15.1 and 15.2.  It is possible that I made a small change to these patches (output volume) but don't quite remember.  Again, returning everything to the preset settings in the manual got rid of the problem.

I am curious to know if those who have many custom presets have encountered  problems with these only, or if any other glitches to factory presets have been identified…or perhaps my exprience was unique.

BTW, the fine-tuning and input meter features are great!

Dave

[timothyhill]

I seem to remember that several people had similar problems with the V3b1 update. Basically, the values stored in the presets don't change, but the parameters they control do (which has essentially the same result).

Thanks for letting us know which presets have been affected, too!! That way, we'll know what to look out for, so if you find any more, please continue to post them. I'm sure there are a few more, and, as your experience relates, they can be corrected with a few minor adjustments.

[aceshigh025]

I already emailed support about my issues, but I thought I'd repost them here too:

Update problems using Snow Leopard:

I had some issues trying to update where it would freeze at the "getting ready for update" part, and I tried all of Eventides tips numerous times. I finally got it to work by letting the Update Utility start the update and put it in update mode, which then led to the freeze, I then cancelled it, but left the Timefactor in update mode, quit Update Utility, and went through the update process again and it went fine, until the very end when it froze at the last step "restarting the timefactor" or something. I then cancelled the update and the newest software was on my TF from looking at the TF's menu. So it looks like it went through, but it never "fully" completed it. For some reason if I manually put the TF in update mode Update Utility won't read it, but If I let the update process do it and restart the process, it works.

Bugs in new software:

The feedback knob on the tape delay says "slur" on the display but still controls the feedback, and the level meter only works 1/4 of the time.  I have to keep going in and out of it till it works.  I don't know if this was caused by the freeze at the end of the update that I encountered (that freeze happened last time I updated to a different software as well, about 3 months ago), or if it really is a bug in the software.

[wedelich]

Hi aceshigh025,

What version of the Eventide Update Utility (EUU) are you using?  If you're not using the latest v1.6, you should download it from any factor's support page. 

[aceshigh025]

I am using the newest version 1.6

I've gone through every inch of this support site to work through the issues I described and spent hours working with my Timefactor to get it to update properly.

[timothyhill]

Aceshigh025, what do you have MIDI CLK OUT set to? Having this setting turned ON can cause the update to hang when EUU tries to reboot the pedal, almost sounds exactly what your last result was.

I also had a separate thread on the Slur function seeming to still control FeedbackB.

[aceshigh025]

The MIDI CLK OUT is set to OFF on my TF.

I'm pretty bummed about the Slur not working. I was really excited to see that as an update.

[brock]

Quote:

I am curious to know if those who have many custom presets have encountered  problems with these only, or if any other glitches to factory presets have been identified…or perhaps my exprience was unique.

I'm just getting back into this, but I'm uncovering more with the PitchFactor back in its usual signal chain.  The custom presets that I left in are indeed changed.  (Can't say that we were warned to backup before the updating process).   That seems (at first glance) to surround the Depth / Key and Speed / Scale settings.  For example, presets that I know were in E min are now in C min (reset to default key?).  But that wouldn't explain the occasional C DOR that I find, or C# min.

And it's not limited to the two diatonic modes.  I have a custom PEGGIATR preset that reads ATK-274.  A turn of the knob (CATCHUP ON) brings back the -10 to 10 range.  Weird.  I'm not getting the same squeals and level problems that you report (so far), but there are changes to my custom presets.  I'm going to see what sticks when I manually restore those presets, and / or restore them from my third-party .syx files.

The good news is that my MIDI assignments appear to be holding now.  If the default CCs have changed from the original UM documentation (for whatever reason), perhaps we should update those changes in some supplementary docs.  MIDI channels reverting to Channel 1 … Well, I expected that to happen.

Quote:

I was able to rename presets in FactorLib by double-clicking on the preset name. Again, that's under Win7, so I'm not sure what WinXP will do.

I should've been clearer, Tim, but I was posting as the updating process was developing.  I can change the the preset name in FactorLib, but it wasn't transferred over to the PF during the upload.  (Custom presets kept the Bank : Preset convention.)   Again, I haven't used it enough to draw any definitive conclusions.  But it is disconcerting to find the Edit Console selection to be greyed out (no matter the selection or situation).

I have a variety of O/S to choose from here, but my 'beta testing box' has WinXP.  I start at the 'bottom' (oldest O/S and computer – not much to lose), and work my way on up.  Let's see what happens in a few other systems, with the 'Factor pedal in its most-used audio signal chain and MIDI routing.

[brock]

Quote:

(Can't say that we were warned to backup before the updating process).

Oops.  Big typo there, and no post editing available.    "I can't say that we were NOT warned to backup before updating."  It's clearly stated, and standard practice.  I purposefully left some custom presets in to find out what happened, and I did have some backups (untested .syx and a mental snapshot of the critical parameters) before I began the update.

[timothyhill]

Brock, yeah, I think this may be a difference between versions of Windows or maybe the Factor pedals. I haven't tried renaming PF presets in FactorLib yet, but they transfer correctly to the TF. The new name didn't appear until I changes presets once on the pedal, but they were there after that. They do have the Bank:Preset convention, though.

The Edit Console works for me with both the PF and TF. That is rather disconcerting that it's always greyed out on your system. It seems to always be available on mine as long as I have a Factor pedal hooked up, even if I don't have a file open in FactorLib. If I don't have a file open and a preset selected (so that it turns yellow), the "Get" button is gone, but I can still open the console from the View menu.

I have found that MIDI CLK OUT needs to be OFF when using FactorLib. That's kind of a PITA, but I can deal with it.

That's not terribly surprising about the changes to Depth/Key and Speed/Scale, but that doesn't make it any less of an inconvenience.

I've got more time tonight, so I should have some experiences to relate with the MF and Space later…

[timothyhill]

Quick update… I have successfully used FactorLib 1.8 (the new version) with all 4 Factor pedals to read preset data from the pedal, edit it on the PC, and send it back to the pedal. The only thing I've noticed is that MIDI CLK OUT needs to be OFF (just like it does with the Eventide Update Utility). For reference, I'm using Windows 7. I have not experimented extensively with the Edit Console, but have had some success with it (TF).

I've also noticed a change to the ModFactor's Vibrato algorithm, specifcally the filter stages (x-knob) in the retro type.

I've also notice that the display in both the TF & MF (Brock noted this in the PF earlier) sometimes displays "weird" characters (Brock noted "ATK-274") in some (not all) parameters. The display reverts back to normal ("expected") values after turning the knob. I can't replicate this, though, after it happens once (in any given algorithm-parameter combination), it doesn't happen again. It's strange, it doesn't seem to affect anything, but it wasn't happening before the V3b16 update and it seems to go away after the first edit. I haven't been able to replicate it and I doubt it's anything serious.

Testing continues…

[timothyhill]

Aceshigh025, not sure if you noticed, but Dan Gillespie from Eventide posted a reply to the question about the new Slur control in the Tape Delay in the "TF V3beta16 Tape Echo Slur Control" thread. Basically, it is still controlling Feedback B and is just erroneously labeled. It should be returning to the correct label in the next version.

[daggodag]

I had some time today to play some more with the new software and may have come up with a potential stable solution to upgrading to 3.16 safely.  Factorlib 1.8 Does work with version 3.2, so download it first, then save your presets.  Following that, update to 3.16, then re-load your presets to the pitchfactor.  By doing it this way, the artifacts don't appear to show up on user presets as they did earlier when I I upgraded directly to the pitchfactor "over" my user presets.

Dave

[brock]

I had an opportunity to download a fresh copy of FactorLib 1.8 onto an OSX laptop, and connect it to the PitchFactor.  So that's what the Edit Console looks like!  I had a go with the renaming process for custom presets.  Still no joy, although it did work on renaming the remaining factory presets.  As reported elsewhere, I had to switch away from the sent preset before the renaming became visible.  But I did find a workaround for custom preset renaming:

If I named a custom preset *anything at all* (Preset 1:1 named "Z", for example) from the PitchFactor itself, saved it, then brought it into FactorLib for more text editing, the returned custom preset's renaming 'stuck' from there on out.  Transferring all 100 presets back & forth was (subjectively) more reliable than sending / receiving one single custom preset.  Occasionally, I'll "get" two complete uploads (100-preset untitled .syx files) from a single Open From MIDI command.

Here's where the process diverged from the WinXP – SP3 experiments earlier.  When I uploaded all 100 presets (mixed custom & factory) to the FactorLib, and Save to Pedal immediately, I ended up with those "weird characters" reported earlier here.  Only on custom presets, and on every one of those (custom).  It seems to be some variation on program change number, plus some value.  By way of example:  1:1 was renamed [2] 1 3.  1:2 was named [3] 1 3.  2:1 was renamed [4] 7 3, and 50:1 became [100] 3 3.

Stranger still: Two or three consecutive full dumps to FactorLib resulted in 114 presets!  The "extra" 14 presets were all named "Preset 1:0".  Just to prove to myself that this was no illusion, I performed a Select All and tried to send everything back to the PitchFactor.  The warning message "Cannot send more than 100 presets" popped up.  The third time, I watched the PitchFactor display.  Only 100 presets were sent, but 114 presets were "received".

So, moderate success in OSX, and slightly less in WinXP.  Note that the Eventide Update Utility would not work for me without SP3 installed.  Now, on to the next O/S …  For those keeping score, an informal sampling of forum responses has Win7 users reporting the  minimum amount of "anomalies".   But the issues here deal mainly with custom presets, and I don't know if the Win7 users were 'all factory', or a mixture of custom / factory presets.

Definitely there are changes made to any number of parameters in a custom preset.  I'm trying to determine whether that's FactorLib 1.8, PitchFactor's beta[16], the host operating system, or some combination of the three.

[nickrose]

Hi Brock -thanks for a good effort in reporting your problems. But, we cannot duplicate any of it. Presumably there are differences between your setup and ours. So, some tedious questions:

1) What version of OSX ?

2) What version of PF ? You imply 3.0.0[16], but please confirm.

3) Did you either do a [CLEAR SETUP] or a restore Factory Presets after upgrading ?

4) If no in (3), what version did you use to create your custom presets ?

My guess is that all will be fine if you use the latest version of PF and all presets were created by it. There may be issues if you mix presets from one version and software from another. Once we get a clear recipe for this problem we can look into it. Such a problem would probably apply to all of OSX and Win7 and XP.

Note that the UM says that "FactorLib requires Windows XP, service pack 3", so we feel no shame if it does not work without SP3.

.

[brock]

I don't want to leave the impression that these reports are real "problems" for me, Nick. It's a public beta, and my goal is to create / share custom presets that will work across most any user's system.  The answers to your questions:

1)  Mac OSX 10.5.8

2)  3.0.0beta[16].  FactorLib 1.8.

3)  No, and no.

4)  Majority created in 3.0.0beta[0]; a smattering created in 2.3.0[2] (original version at purchase); very few remaining 'factory' presets.

Which brings us to your next comment.  That makes perfect sense; especially if there were radical adjustments made to the code.   I didn't get that kind of result in the jump from 2.3.0[2] to 3.0.0beta[0].  Regarding the WinXP SP3 requirement, I missed that key 'factor' in 3.0.0beta[0] / FactorLib 1.x documentation.  I thought it would be worth repeating here.

There's nothing happening in the 3.0.0beta[16] transition so far that I can't repair, or circumvent .  But if *whatever it is* can get quantified, so much the better for the next iteration.  If you don't mind, Nick, I have a couple of quick questions for you:

1)  I seem to remember a tidbit about having to leave your stompbox at the default System -> MIDI -> SYS ID 1 to communicate with FactorLib.  Holds true for FactorLib 1.8?

2).  On the PitchFactor' end, while uploading a single preset:  Does the MIDI IN/USB port always accept an incoming preset .syx message, or is it 'listening'  for a 'handshake' / sysex header of some kind first?

In the interim (3.0.0beta[0]), I toyed around with a favorite freeware app (Bome's SendSX).  I could send and save a single preset, the whole dump, system settings, etc. TO software, but I never had any success in sending it back to the PF.  Yet, those same saved presets load just fine into FactorLib 1.8.  I tried quite a few combinations the PF's system settings, and scoured your online sysex information (to no avail)..

[fkalich]

You guys should be happy, considering my horror story.  I decided to upgrade my 3 eventide stomp boxes.  I am always afraid to do upgrades, I had only upgraded the Pitchfactor to the early beta(0) up to today.  Well the Timefactor and Modfactor went well, but my Ptichfactor failed to upgrade, the utility told me to try again.  I did, again and again.  I tried a different computer, different cables, everything they suggest and more.  But now I am a Pitchfactor that does not work, and will not upgrade.  I worked fine yesterday.  I have contacted them by email, but unless someone knows of some secret remedy (I figure near 0% probability) I am going to have to send it back.  Fortunately it is in warranty, but I am not a happy camper over all this.  And am doubtful that I would attempt any upgrade in the future when I am out of warranty, after this experience.  I would think it was a defective unit, except that it worked fine before the upgrade attempt.  Any miracles that anyone can pull out of their hat here?

[timothyhill]

Can you provide more information? For example, what are the specific error messages you're getting? What type of computer/operating-system are you using? What version of the Eventide Update Utility are you using? Have you tried resetting the PF to the factory default settings? Do you have midi clock out turned off? Basically, the more you can tell us about your system and what's happening, the better the chance someone here can help you.

[fkalich]

Thanks for your quick response Tim. Here you go:

1) The Midi clock on this unit was on, I used it as master for the other two units.  I don't recall seeing anything in the instructions saying to turn that off before upgrading.  However I am not able to enter system mode to change anything now, at least not in your prescribed fashion.  I can only go into update mode using the middle switch.  Any other combination of switches (such as right foot and encoder) on start up result in the lights turning on for a second, and then turning off, the unit going dead.

2) I get this error when trying to update(or something very similar).  It occurs in the write phase.

Error Updating Device    Please reboot your device and try updating again

The error occurs at different % complete on every try.  Once it got to 70%, but generally it fails at anywhere from 10% to 30% complete.

3) I am using the version 1.6 Build 13 of the utility, the most recent I believe, I just downloaded it.

4) I have tried two different computers, both Windows XP Professional, both up to date.  I tried different USB cables.  I disconnected all other USB devices, and shut down all other programs, everything you instructed.

5) My computers are not the highest powered or newest.  However I had no problems upgrading the firmware on my Timefactor and Modfactor today.

Thank you:

Frank

[jhyman]

Hi Frank,

Please take a look at the email that I just sent you. It contains an alternate method that might help you out. Email me back with the results.

Jerome

[fkalich]

Thanks for prompt response Jerome.  I tried the other method of updating that you sent me, I got this message during the write phase of update

bad flash write send data operation error 3

To me this looks like hardware, but I am no expert.  I did mention that the Midi clock is turned on, but I don't know how to turn it off (as the only mode I can enter now is update mode).  If you know of a different way to get to system mode than middle switch/encoder (which does not do the job now), I can try turning the Midi clock off.  However that may not be relevant now, you can tell me based on the error that I got.

I am not an unhappy camper now. I mean, things are screwed up a present, but you guys are trying your best to help me.

Frank

[fkalich]

I thought to add something for the benefit of others who might have the problem that I have had.  As I mentioned, I had no problems updating the firmware on my Modfactor or Timefactor today.  I finally got the Pitchfactor to succeed in an update on the 20th or so try.  I gave it a rest after about 19 or so failures, and it did then get all the way though the update firmware write on the last attempt. On the prior failed attempts, I would get a "flash write memory error" at varying %'s of completion.   Jerome thought that perhaps an electrical spike on my first update attempt caused the problem.  Everything works now, but I want to do an upgrade again before my warranty runs out, to assure myself that this was only a random occurrence, and not a hardware issue.

[timothyhill]

Hey, Frank! I'm glad to hear you got it working again. Jerome may have been correct, or there may have been a number of elements that just came together at the wrong time.

From your second post, to me, it sounded a lot like behaviors I've seen when MIDI CLK OUT is ON. The way it's been explained to me is that some early USB 2.0 controllers (part of the computer) just can't handle the clock messages. That causes them to either empty or corrupt their message buffers and the end result is that the EUU loses contact with the pedal and the update dies. Unfortunately, that *is* a hardware problem with the USB controller (not with any of Eventide's hardware), and the only way to fix it is either getting a new USB controller or a new computer.

Fortunately, there is a simple way around it… just make sure you turn off MIDI CLK OUT before attempting an update. That applies to all factors, not just PF, and I've seen it happen with FactorLib, too, albeit with less-catastrophic results (they just won't talk to each other in this case).

That's not to say that Jerome doesn't have a point. A power spike or fluctuation, a short in a cable, a bit of oxidation in the USB jacks or plugs, any of that could have potentially occurred.

I'm just glad, as I'm sure you are, that your PF is once again functional. Just remember to check the MIDI CLK setting next time and maybe look for a less stormy day. (I have no idea where you are, but where I am, we've been getting pounded by strong storms and tornados all night. Ah, just joys of spring… but I digress).

[brock]

Long overdue followup here.  I immediately downloaded Version 3.0.0[19] when it came up, and used my original WinXP setup to facilitate it (described in detail above).  Smooth as silk installation.  Nice work, ET.

OK, still the known issues on ver. 2.x presets and the "Preset 1.0" phenomenon.  I expected that part, and used the workarounds.  Certainly no showstoppers there.  But the exact same setup that I'd used previously (plus a few minor XP updates, I'd guess), yielded great results with 3.0.0[19].  The FactorLib Edit Console works, no weird parameter changes or characters, etc.

I am zeroing in on a Harpeggiator issue with significant rate / tempo division changes (and a possible MIDI clock, or Tempo On/Off connection), but I'll save much of that for another thread.  I don't have a 100% recipe for it yet, but I've experienced a little weirdness here at least back to 3.0.0beta[16].

Rate changes controlled by an EXP pedal.   Seems as though the PF gets stuck at 500 BPM or 10ms., and only a tap of the RESTART gets it back to the current value.  If / when it happens, you'll know it (especially evident when the filters start 'gargling').  Sometimes, a (custom) preset will load at this extremely rapid rate.  RESTART, and it behaves as expected.

[Author]

Posts