Problem with Beta 6

[cici]

I don't know what you changed to get it this good. But if you shifted it 70ms, I'd suggest trying an extra 7ms shift. Similarly if you shifted it 200ms, I might try an extra 15 or 20ms.

Igor,

When testing my "COLOURS" project in beta8 V, I noticed a very small delay for just a single "cut" transition. I suppose that Harold's suggestion will be the right one.

Thank you very much

[Igor]

1) Is this delay for one slide or for all slides?

2) How it will work in v3.80 or v4.01?

3) Please make sure again that you really re-compiled the show in v4.10 beta #8. You can compare with beta #7 where there is a delay in 100 ms.

[HaroldB]

Igor,

Let's see if I can try to explain this so that it makes sense .

When you have a show with quick transitions set to music with a quick beat, sometimes you hit the EXACT right spot in the music with the transition.

I find that the new beta 8 works just fine for almost all of my slide shows. There is the one I sent you which is just a tiny bit off from where 4.0 was. I imagine that it might be something incredibly small like 10ms or so, which is why I don't it at all "see it" with my other shows. But this show just seems to be oh-so-slightly off in that one or two of those previously "exactly right" transitions are no longer "exactly right".

It's possible that the whole show might be off by that 10ms, but for almost all transitions, 10ms is too small to "see". But when doing sharp cuts, sometimes you CAN see it. So it appears that only one or two slides are "off" when really they all are.

Beta 8 is good enough that I can live with it. But maybe you can go back and check your calculations just one more time to see if it might not still be off by some tiny bit? It certainly might be me, that I am watching this slide show over and over again while I should be sleeping .

Respectfully,

Harold

[Igor]

I just prepared two new test versions. Harold and Cici please try out:

http://www.wnsoft.com/apr_beta_FastTimer_100ms.zip

http://www.wnsoft.com/apr_beta_FastTimer_50ms.zip

Both variants in 2 times more often check up current position of playback (even much better than v4.01 or 3.80).

First variant has 100 ms. time compensation (as Harold tested with beta #8). Second has 50 ms. time compensation.

Which variant will provide more exact timing as in v4.01?

p.s. interesting moment. Compare how smoothly (in both betas) cursor runs on the time-line under Windows 98/Me than earlier before.

[cici]

I made a test with my “COLOURS” project as I used in this show many cuts.

I tested it with beta 6 and beta7.

In Preview mode I noticed some delay during many “cuts” transitions.

When viewing in time line (Little window) I noticed that the whole music track (MP3 /128 Kb sec) was playing just about 2.5/10th seconds ahead!

Of course, in “preview “it is difficult to notice the small delay during the long “fade in/out” transitions, but it exists. And just because of the regularity of the delay, some “cut” transitions appear to be synched at the right time but they are not.

I made a test with the “beta 7 version with old way of running Preview: the result is much better but not perfect (referred to the original project).

Igor, I completely agree with Harold's explanation about the very, very small delay I ''can'' notice about just one ''cut'' transition and I think that following his last suggestion will put everything at the right place.

Anyway, the difference is so small that we can consider beta8 version as OK and I think this difference is noticed just in Harold's show and mine which look like to have a very special timing.

[HaroldB]

Which variant will provide more exact timing as in v4.01?

Igor,

The 100ms fast timer version is better.

I think the 50ms version went in the wrong direction -- iow, it made things worse, not better. I don't know what exactly we are measuring here in milliseconds, but could I trouble you to generate a PTE with 120ms? That's the direction I think might solve the problem. The 100ms fast timer version is ALMOST there.

Thanks a lot!

Harold

[Igor]

Ok, I won't change anything in timing more. Will leave as in beta #8

Thank you all for testings!

[HaroldB]

Ok, I won't change anything in timing more. Will leave as in beta #8

Thank you all for testings!

Igor,

I assume that you posted your message about leaving it as in beta 8 before you read my message posted a few minutes earlier .

Even the newer version with 100ms is better than beta 8.

Harold

[Igor]

Just I thoughted that it better to don't change more, because this more frequent query of current position takes much higher CPU load (on 10% under Athlon 1600) in the Cust. synch. window. And possible another untested nuances.

[Guest guru]

Igor, I too checked beta#9 with 50 and 100 ms. compensation, and I too agree with Harold and Cici when they say 100 is better.

And I too want to say, like many others, you are doing a more and more outstanding work. Thanks, Igor.

[HaroldB]

Just I thoughted that it better to don't change more, because this more frequent query of current position takes much higher CPU load (on 10% under Athlon 1600) in the Cust. synch. window. And possible another untested nuances.

Igor,

Maybe you should just try without more frequent querying of the CPU, but see if a 120ms adjustment (instead of 100) makes it better?

Harold

[Igor]

After I've added that time correction (100 ms.) in the beta #8 with your help, I see no difference between v4.10 and v4.01/3.80 more.

Just for test you can try move several transition to the left on 20 ms. It will not take visual difference.

[alrobin]

In CoolEdit, I discovered that each time I save an MP3 file, it adds .027 seconds of silence to the beginning of the file.

Harold,

I find that if I add 3 or 4 seconds of silence to the beginning of an mp3 file in Cool Edit and save it, then when I reopen it, most of the silence has disappeared. The silence in the original saved file, however, does show up in PTE. Have you experienced the same effect?

By the way, speaking of Cool Edit, as you probably know, Adobe have bought Cool Edit Pro, and will be re-releasing it this fall as Adobe "Audition" for about half of what it costs now to upgrade to Cool Edit Pro. This offer only good for a limited time to current owners of 2000 or Pro.

[cici]

Hi Igor, Hi Harold,

I’m not an expert at all but

I don’t understand why, as we found (both Harold and I) the delay is a constant delay of about 0,25 sec (0,25 sec for me and 0,27 for Harold)), the correction is of only 100 ms! In my modest opinion the right solution is a correction of 250 – 260 ms.

Of course a correction of 100ms is much better than nothing ; a correction of 120 ms will be still better… for some slides, but there could be a bigger problem for others slides.

Anyway , we are sure that a correction of 100 – 120 ms will not solve the problem and if this correction is dangerous “because this more frequent query of current position takes much higher CPU load”, the decision to get beta 8 unchanged is the wisest one.

I’ve no problem with (better: I can’t notice the difference in) others shows than COLOURS,

but this show has a succession of quick “cut” transitions falling at definite moments of music.

If I had to make a new exe, just 2 solutions: save version 4.01 or move transitions 0,25 sec to the left. There is a (small) problem, but you can notice it just in a very particular case. We can say there is no problem.

[Igor]

Anyway, in the new presentations will be no problem at all! Because we already create timing in the new music player

[alrobin]

Sam,

For sure, if you have plans to go back and change a previously-created show, you should keep the version of PTE in which it was created. As you probably know, you can have as many different versions on your desktop as you wish.

I don't think it is reasonable to expect Igor to keep the operating characteristics of each new version exactly the same as each prior version - else, how would we ever get to enjoy all the new features and improvements that we have requested?

It's impossible to add new features and not upset somehow, and by at least a small amount, the way PTE operates. All we can count on is that each new version should create new shows better, faster, easier, more precisely, and with more artistic capabilities than the previous version.

[cici]

Al,

I agree with you and my last topic wasn’t aiming to claim what I was expecting, but to support Igor’s decision. For sure my expression didn’t follow my wish and I’m sorry for this.

As you, I’m only interested in “ new version” to “create new shows better…”

This difference in timing isn’t a problem at all for me. But what is not a problem for me, can be a problem for somebody else and he has the right to ask for a solution. Actually I think that a bigger difference could have been considered as a problem by professionals working with a template, and it is in that view I decided to make some tests too, as Igor had shown his will to solve this “problem”.

If there is something to put in evidence, I think, it is not the reasonableness or not of a request, but Igor’s receptiveness: he had the right to give the same answer as yours, but he didn’t. Even more, he tried, hardly tried to give a solution to a very small problem, and this is a sign of a big kindness, intelligence, wisdom.

Friendly

[alrobin]

Sorry, Sam, my comments in the last 2 paragraphs weren't meant to be directed specifically to you, but to everyone in general, and particularly to anyone who might be frustrated that the parameters for the final 4.10 beta might not exactly match up exactly with previous shows.

I was trying to be philosophical, but I guess as usual I came on a little too strong. I apologize.

But, I'm glad you agree with me.

And, I agree with you that Igor is bending over backwards to modify PTE in the most acceptable way in order to be able to give us more control over synchronized presentations.

[HaroldB]

Just for test you can try move several transition to the left on 20 ms. It will not take visual difference.

Igor,

You are perfectly correct that I could not see any visual difference when I manually shifted everything 20ms.

But then I tried the test with shifting everything 50ms. And that seemed almost exactly like the older releases!

I think you should increase the factor of 100ms to 150ms, and that will solve this problem. Of course, it may aggravate other shows, and then I'll have to be embarrassed and have to ask you to change it back to 100ms . But I really think that 150ms will do the trick.

Thanks for working on this!

Harold

[HaroldB]

I find that if I add 3 or 4 seconds of silence to the beginning of an mp3 file in Cool Edit and save it, then when I reopen it, most of the silence has disappeared.     The silence in the original saved file, however, does show up in PTE.    Have you experienced the same effect?

Al,

I don't recall that ever happening to me. I just ran a test and it worked just fine, too. I am using Cool Edit 2000 1.1 build 2418 -- is that what you are using?

By the way, speaking of Cool Edit, as you probably know, Adobe have bought Cool Edit Pro, and will be re-releasing it this fall as Adobe "Audition" for about half of what it costs now to upgrade to Cool Edit Pro.    This offer only good for a limited time to current owners of 2000 or Pro.

I got an email from them.

Have you ever used Cool Edit Pro? Does it offer everything the plugins for CoolEdit 2000 offer?

Harold

[HaroldB]

I don't think it is reasonable to expect Igor to keep the operating characteristics of each new version exactly the same as each prior version - else, how would we ever get to enjoy all the new features and improvements that we have requested?    

It's impossible to add new features and not upset somehow, and by at least a small amount, the way PTE operates.

Al,

Actually, speaking purely philosophically, I disagree with you, at least a little .

While I realize that it is necessary for some old shows to occasionally break with new releases of PTE, I think a design goal of PTE (and for that matter, all software) should be for that never to happen.

After all, think how Igor would feel if Microsoft kept on changing the compilers he used so that with each new release of the compiler, he'd have to tweak the code before it would again compile!

I DO realize that it can happen sometimes, preferably for rarely-used features. For example, in the last year or so, Igor got rid of a particular transition I used, and I went back and changed my shows accordingly. With the beta versions of 4.10, I had one show that suddenly crossed that "tight transition time threshold" and I had to go back and change it.

But I think that such incompatibility should be kept as small as possible. And it should only occur when necessary. In this case, it is simply a matter of finding the right adjustment factor to make the new version of PTE work properly -- there is no real reason to leave an incompatibility.

Here is another factor for the case in point. Up until now, the timings in Cool Edit (and I assume other sound editors) and PTE matched perfectly. It makes a lot of sense to me to keep it that way.

I don't think that you and I disagree much on this. I think that I'm just articulating a slightly different emphasis.

Harold

[think(box)]

Harold, Sam, Al - I would like to thrown in my 2 cents: If P2E show is inaccurate (bad timing) or incorrect (functional error) or slow in performance (e.g. load time) then the change is welcomed! I do not want things to stay as before under those circumstances. Most of this topic thread is related to inaccuracy. Igor's endeavor to achieve the best tradeoffs is just perfect.

That all sounds more or less like violent agreement with everyone, with a healthy dose of gratitude to Igor.

Regarding Microsoft changing things on us: Look at what happens to our user interface and operating environment every time they release a new version of Windoze. I think Igor should be applauded for how he has resisted reinventing the world with each new PTE release. Unlike as with Microsoft, we do not have to relearn the basics as each new version of PTE is released. In Microsoft's defense though, their software is all about a "user interface" and the "engine under the hood". How can they improve it without driving us to distraction? XP's yesteryear mode (classic Windoze interface) is there for just that reason.

[alrobin]

After all, think how Igor would feel if Microsoft kept on changing the compilers he used ......

Hi, Harold,

Microsoft do that to me all the time! (my Cubase MIDI sequencer won't run on my laptop under XP, Visual Basic changed courses a few years ago so I can't even run some of my old programs, VBA changed from ADO to DAO (or vice versa), screwing up some of my earlier developments, I can't even save Adjustor in Excel 2000 - have to go back into Excel 97 and save it, even for use under 2000 ???, etc, etc.)

However, you are right, ideally and philosophically! The world should be a better place! All software should be backwards-compatible, and all new forms of entertainment players should be compatible with all previous formats (including beta videotapes and 8-track tapes! (I'm Irish - I'm allowed to mix my metaphors! )

But, ideally speaking, I agree that it would be great if new PTE versions are kept as compatible as humanly possible with previous versions, and still continue to progress in terms of new features and ease of operation. As a design goal, I agree with you 100%. I guess I was being more pragmatic than philosophical!

Re Cool Edit, yes, 1.1 - 2418 is what I am using, too. I played around with the "Pro" on a free trial basis, and it was pretty impressive. It will rip from a CD, which 2000 won't let me do (at least on my setup), and it will also detach a sound file from an avi file, which is something I had to do to get the sound for one of my shows, and had to have a friend do it with his video software. It has something like 120 different tracks for mixing sounds and music, which could be handy, and a lot more special effects than 2000. I've put my order in for the new "Audition", due out this fall, so will let you know more about it when I get it.

It would be nice to have PTE match other sound editors. But, I thought from reading the postings from you and Sam and Igor that the problem was more one of compatibility with previous versions of PTE, and the way PTE reacts to the music, not a difference between Cool Edit and PTE operation. But, maybe I don't understand the issue completely (that's what happens when you take a few days vacation! )

The problem I have with Cool Edit dropping the silence at the front of a piece occurs with an mp3 file which is the result of a mix-down of several .wav files and when I re-load it into "2000" to make an alteration. So, in order to maintain the timing of the music, I have to go back into the original "session" and make any changes there, and re-mix everything. Probably the proper way to do it, anyway.

[alrobin]

Well put, Bill!

Man, this thread is getting interesting! Sounds like we are all on virtually the same wavelength on this.

Meanwhile, Igor is back in the lab, slaving away to try to make us all happy! Wish we could get Bill Gates' attention this easily!

[LumenLux]

Ok boys, "dumb this down" a little for some of us. Maybe I have not followed closely enough on this, but here is my question. What is the "backward compatibility" general concern? Is it only when re-working an earlier show? I mean, once we have the PTE.exe file, we never need any concern with Igor's changes effecting the playability of that particular show, right?