AQ updates SCAMP to 1.1. NOW 3X FASTER!!! Evaluation Reset!

[AdmiralQuality]

I had said I was going to hold off on this until Poly-Ana was finished, but the new filter optimizations from Poly were working SO well that I just had to get it out there. So I took a couple days off from Poly's never ending slog to spend on updating SCAMP...



http://www.admiralquality.com/products/SCAMP/

New in SCAMP 1.1:

- New code and compiler optimizations have greatly improved performance without affecting sound. SCAMP 1.1 now uses less than one third of the processing power required by version 1.0!
- Anti-denormal technique modified to further improve performance.
- Editor now loads faster.
- Updated background and about-box graphics. Background labels should look a bit sharper now.
- Added 2 new presets including a "plain vanilla" LPF default and tweaked settings of original presets.
- Bypass buttons now automatically pop out to indicate channels are unconnected.
- Bug in program save of bypass buttons fixed.
- Free evaluation period reset, including for those who's pervious SCAMP 1.0 trial has expired.
- Added a wet/dry mix control. New .ini file ID is MIDICCWETDRYMIX (or just delete or rename your existing SCAMP.ini file to automatically re-create the defaults).
- Knobs now zoom to 8X in linear mode when SHIFT key is held. (Was 1.5X before.)
- Fixed an bug caused by uninitialized MIDI Mod scale until the first controller movement.

That's right, even if you've already evaluated 1.0 you can evaluate SCAMP 1.1 again free for another 7 days. So TRY IT AGAIN, FOR THE FIRST TIME.

I should note that SCAMP's introductory price offer of $39.95 is expiring June 1, 2006. So try it soon and buy before June to save $30!

Now back to the synth... (slog, slog, slog...)

[diverdee]

I for one will definately d/l & try this again.
3 x faster, hefty optimization

[.jon]

Excellent update, demo reloaded

[ObsoleteAcc99]

[DELETED]

[ObsoleteAcc99]

[DELETED]

[AdmiralQuality]

ok. well I am happy enough that a purchase is imminent. Thanks for listening:) I am not seeing less than a third of cpu usage, not by a mile. Live gives it in precise numbers and it's exactly half of what it was before.

What CPU are you on ttoz? Also remember if bypass buttons 3-4 show as in now, then you're probably wasting power.

I get 1/3rd the load (even a bit less) on my P4 3GHz @ 533.

I rescanned lives plugin folder (complete refresh) and it temporarily fixed the "invalid" problem (more on that later)..

Anyway, some observations:

If the outputs 3-6 are physically disabled in the ini file, it results in 1% less cpu usage (4% vs 5%) then by bypassing those 4 channels from the scamp gui.

Good observation, yes it does (though I'd say the percent of total is a bit less than 1/5th) Bypass is a bit more expensive than just limiting the NUMCHANNELS. (It needs to make a decision to bypasss and still has some work to do, for one thing it's collecting levels for the level meters when the channels are connected, but bypassed.)

All 6 channels use 12% which makes sense. This speed improvement is much less dramatic, as that's what it was using before the 1.1 update with all channels enabled. Not even half, nowhere near it, and definitely not a third!

Thanks for the report. I'll look into it further when I find out your CPU make.

However, 4% I can handle for a stereo instance for a filter of this quality. AND it fluctuates between 2 and 3% with channel 1 only, which will be nice for some mono filtering goodness.

I'd like to just inform that nothing has changed in the project or my latency or my system since scamp v 1.0. I have a project set up simply called "tester" which is a single audio track and single midi track. I put an audio file on the audio track and leave the midi track empty unless I want to test a synth's cpu usage. Anyway so of course in scamp's case I simply test with one audio track. It's a standardized test i use to find the exact cpu usage of a plugin.

Sounds good.

Ok, the problem with scamp not being a valid vst anymore is happening again. i have maybe 150 vst, and only two of them have EVER displayed this behaviour, R66 and scamp. R66 also used to have problems in Tracktion where a project with a saved r66 would reload and say the R66 was missing. A cache clean and rescan would temporarily fix this, as it does with scamp....and the developers released a beta to fix this with R66 which I have been told workd (haven't tested yet), so i DO believe the problem is definitely on your end.

Ug, Tracktion. Yeah, it's my fault, sure. Works fine in every other host but I introduced an anti-Tracktion bug. Seriously though, will try to find out more on this. Can't find anyone who can reproduce your problem however. What version of Tracktion do you use?

Also, one more question.
When I start with cutoff at 60 as i start sweeping the sound momentarily softens then goes back to the original volume (Low pass)..it's actually better to start sweeping from around 20 (3/4) ..is this normal? I understand the audio will be noweher near 60khz but it's just sometimes common practise with a filter plugin to sweep from max to min and vice versa....

You'll probably find this filter has some extra range on the knob compared to most others. Anything clockwise of 12:00 is pretty much beyond your project's nyquist (but the knob has to go there as the LFO and Env Follower interact with it, so you have to be able to go beyond normally useful ranges.) But you should find the cutoff freq is close to the number reported. Just sweep through the range you want. Also note if you're using a MIDI CC then you have the range knob for adjusting it so you don't waste any controller movement. Likewise with the ranges and polarites that can be set in the .ini file for every MIDI driven control.

Also, a major feature request. clicking on a know after you have gone to another knob, it's impossible to get the precise point of where the setting was judt to make some fine tuning. it always goes where you first click the mouse. There should be a way that you can click and the slider/knob doesn't jump..make sense??

You want relative movement. Yes. Someday. (Poly needs it so the code is coming) So sure, relative circular mode in the 1.x series. Promise. (Here I go again!)

...

Oh also, what is your auth policy..is it easy to obtain reauth of one formats a hard drive? And if (god forbid) you were ever to shut shop would you provide your users with a version that doesn't need to be authorized?

It's per-machine. Our pricing reflects that. (Buy a SCAMP for every room in the house!) Still, we are sympathetic to real world problems/machine changes and will in some cases (and entirely at our discretion) issue new keys. Still, we have to stop people from giving away copies to their friend(s) too, so I can't PROMISE this, but I'm poor like you and I understand. So email me and state your case. K?

As far as shutting shop, it's such a small shop that it can't really be "shut", so that would only happen if I died, in which case I'm sad to say I won't be able to help any of you. (But I'll see you in hell!!!! ) ...but maybe if we sell enough copies I can pay someone to take care of this after I'm gone!

I went with least obtrusive copy protection model I could, short of giving it away free. The downside is the machine dependence, but again, we factor that in on the price.

Lastly, can I have my laptop and my desktop authorized? (obviously they wont be used at the same time)

Strictly, no. (But see above.)

[ObsoleteAcc99]

[DELETED]

[AdmiralQuality]

I can't discuss the copy protection any further here (or elsewhere) ttoz. Some changes will change the key, some won't. If something bad happens you need to write a letter to me.

Believe me, I'm not in the business of trying to sell people the same piece of software twice. (Unlike most of this industry!) My customers are the good guys. Looking forward to having you as one.

[AdmiralQuality]

OH...

ttoz, one more time, what CPU do you have? And FSB & RAM speed please? Thanks.

[ObsoleteAcc99]

[DELETED]

[ObsoleteAcc99]

[DELETED]

[AdmiralQuality]

Yeah I'm confused too and I can't get anyone else to replicate it, so it's going to be hard to fix. None of my Ableton testers report this problem with using multiple SCAMPs. CAN ANYONE ELSE WITH ABLETON REPLICATE THIS?

Is this an old project by any chance? Does it happen when you start a fresh one from scratch? (Just thinking maybe it's to do with an old project looking for old SCAMP with one less program parameter.)

Can you find any reliable way to replicate this behavior? You said reloading the project fixed it? Does it always work after a reload and never work during construction? Or is it dependent on current sunspot activity/wind direction?

You didn't include RAM speed in your specs BTW. (I dont care about the spec of the box you haven't built yet! Just the one you tested the speed on and are calling me a liar over. )

Also, can anyone else out there report in on the speed increase (if any). I'm disappointed ttoz isn't seeing the same speed increase I am. I can run 4 SCAMPs at 6 channels each now where previously I could barely do one (so that's actually > 4X improvement on my box!) so the speed increase is VERY noticable to me.

Sorry ttoz, it's not looking good for "fixing" this problem. Let me know if you experience the same thing on the new box. (Betcha don't.)

[Prajch]

Mine is slightly more than 2x faster. Went from ~45% CPU with all 6 channels to ~20%. That's still a pretty huge difference though.

[ObsoleteAcc99]

[DELETED]

[Colonel Flashback]

no problems here with scamp on live 5.2 pc...

will naivelpf also get an performance update?