Happy New Years to everyone! I have read a few questions on the new feature about the benefits and dangers of erasing the "unneeded" +- files. My question is; other than the reduction in HD space used by this files, is there any difference as far as sound quality or generation of the RT's if I choose not to erase them using the new stretching capabilities?
I have also read about the "dont's" on slower XP machines. I'm currently using a Gateway Pentium 4 with 2.6 GHz and 4 GB Ram, is that considered an slow machine for this purpose? I have not had any problems so far on creating my songs on previous versions of BIB. Currently waiting for the DVD's to arrive, can't wait!
Thanks for the answers

Mike B.

Check in the RT settings and see if the 'use +/- files' checkbox is checked or not (might say something different - can't remember exactly what it says). If it isn't checked, you are using the new stretching algorithms. If you want to see how much longer it is taking to generate a song, time one with and without the box checked. If you can live with the difference, erase them if you want to gain some hd space. If you don't need the space, leave them there and if you want the speed back check the box. Nothing to loose by leaving them there except the hd space. Your call pretty much.
Good luck.

My Lenovo of course is fine. The Toshiba is fine with the new file stretcher however I had to create a new hardware profile the disabled all unnecessary stuff such as the network card.

Don't make the decision on speed alone; actually listen for a difference also.
The new way should actually sound better.. at least it has to many of us.

Mike, good advice above. Also, you could run Geekbench on your PC and tell us what you get. It's free at http://www.primatelabs.ca/geekbench/

PG Music did not set a minimum score, but there are a few other known scores out there to compare to.

Thanks for all the answers to all of you. So can I assume that even if I have the +- box checked if I'm not deleting the unnecessary files (space is no an issue at this moment, they are all in an external HD), still the new and old RT,s are going to benefit from the new engine? i have tried a couple of them changing the tempo by quite a difference and even the real drums seem to sound better.

Mike B.

And Matt, here are the scores:
Integer Processor integer performance 1153
Floating Point Processor floating point performance 1041
Memory Memory performance 1040
Stream Memory bandwidth performance 1060

Total score = 1081

Not sure what it all means, but I hope you'll tell me.

Thanks again

Mike B.

Mike, about the +/- checkbox, in Options Preferences RealTracks, there is a checkbox that says "Use +/- RealTracks for Generation". If you UNCHECK this box and everything works fine, sounds good and the Task Manager CPU% does not peak well above 60%, then you could most likely (after a suitable period of testing to be sure) press the button, "Erase unneeded +/- Files". In my case, I did the "Erase" command on a copy of the files, and left the original distribution hard drive untouched, as a backup. Remember, unchecking that checkbox means you are NOT using the "old" style of RealTracks, so they are then superfluous if everything is OK.

About Geekbench, the forum does not have Geekbench reference numbers for machines that work and do not work with the new engine and RealTracks. We probably should start collecting that by anecdotal evidence.

We do know from another post that a PC of about 4 years old vintage with a score of around 1000 was said to work OK. A few others reported scores of 1200 and they did not say the new engine didn't work. Mine is a two-year old Core 2 Duo with a score of 3000. One user of an Intel I5 got a score of almost 6,000. Apparently, the number of cores you have is a multiplier, and BIAB does benefit from using more than one core.

What does it all mean about Geekbench? We won't know for certain until more users on the borderline like yourself report in that it works or does not work without the old +/- files.

One last caveat: if you do erase the old +/- files, you cannot then use older versions of BIAB with that altered RealTracks folder. Versions from 2010.5 and before require those +/- files. I don't know why anyone would want to use an older version after hearing the improvements from 2011. Even if they had an old and slow PC, they could check the checkbox and use the old files with the 2011 version.

Yes I agree on why would anybody go back to an older version of BIB, to me the new stretching capabilities would make me decide in favor of upgrading, it's like having a whole lot of new RT's at my disposal. On the issue about deleting the +/- files (I have them all backed up anyway)I'll probably leave them alone for now until I try the performance as you suggested.

Thanks for your help.

Mike B.

The other thing that made me keep looking was that when you install the new 2011 over the top of the 2010 or 2010.5 etc. Any RTs that are new for 2011 WILL NOT have the +/- files anyway so in essence you'll be using a half&half system. I confirmed this in BETA.

Yes, John is saying the same thing I did in another way, and that's helpful. To put it even differently again, you cannot use the newest 2011 RealTracks in versions of BIAB that are 2010.5 or older, so that's another reason why it would be strange to go back to an earlier version.

While this introduces a compatibility problem for users of earlier versions, it is more than offset by the fact that the program can now be distributed on much smaller media.

And if you choose to use 2011 and keep the +/- files the NEW RTs will have to use the new system anyway as they don't come with +/- files.You have 3 choices:
1. stay with an older version and don't upgrade
2. Upgrade and find a way to make it work with +/- files removed
3. Buy a new computer.
BTW if you don't use BIAB live another option is remove the +/- files but leave "fast generation" unchecked.

Again, thanks to everyone for pitching in on this subject. Very helpful as always.

Mike B.

I have been looking to upgrade my computer and
have been following this post with great interest.

Prior to BIAB 2010.5, without the new stretch feature,
I don't recall experiencing any audio issues.
Since using BIAB 2011, with the new stretch feature,
I started noticing audio drop out.
I tweaked my computer system trying to reduce other background functions that
placed demands on the system, it help a little but occasion audio drop out still would occur.
Using Task Manager to monitor the computer's performance, while using BIAB with the new stretch feature, CPU usage would peak and hang at 100% very often, rarely drop down to the 60 range, but available RAM would remain at 400MB or more.

Testing my system with Geekbench without any other programs running:
Geekbench Score 1510
Processor interger performance 1585
Processor floating point performance 2000
Memory performance 900
Memory bandwidth performance 758
Op. System Microsoft Windows XP Home Edition
Model HP Pavilion 061 EL494AA-ABA a1226n
Processor AMD Athlon 64 Processor 3400+ (1 processor 1 Core 1 Thread)
Memory 1.44 GB 400Mhz
Processor Frequency 2.19
Buss speed 200MHz
L1 Cache 64 KB
L2 Cache 512 KB
L3 Cache 0

I really like the new stretch feature in BIAB 2011,
but I feel my processor speed, Buss speed, L2 cache size, and Ram speed are a little weak to support the demands this new stretch feature demands.
It will be interesting to see other folks Geekbench results and experiences with BIAB 2011 and the new stretch feature.

Bill

Hi DHD,

Bill, I ran pretty much the same system during the beta test as you cite here.

Was able to run everything in BB2011 okay - but I did have to defrag my hard disk first, after loading the new program and realtracks.

You might try the defrag.


--Mac

Hi Mac.

I did defrag when I first started noticing the audio drop outs,
I also eliminated or shut down as much background stuff as I could.

Thank you for your sugesstion.

Bill

Ok, here's what I've done since I was also experiencing some drop-outs and glitches; 1. disabled the fast RT generation thing, 2. unchecked the +/- files usage to try it with older RT's (it works fine)and 3.left checked the high quality regeneration (or something to that effect) when stretching the RT's. So far I have noticed a vast improvement on the elimination of audio drop-outs and glitches. I have not checked the CPU usage afterward, but so far is fine. I'll report later if there are any changes.

Later

Mike B.

Good report. If you are using the +/- files, an interesting test for your machine would be to try the newest RealTracks that do not have the +/- files, such as vibes.

Ok Matt, I try them both ways, with and without it and no problem here with the new RT's with the +/- enabled. kept an eye on the CPU usage, and except for one instance that picked for a second at almost 100%, the rest of the time was below 50%, most of the time actually at around 15-30%. So I guess I'm good for now.

Mike B.