www.dawexpert.com

I would register it and have so much traffic I'd be rolling in the money.

Case in point: me.

Missteps:

CubaseVST5 had a MAJOR conflict with running Gigastudio on the same machine. Workarounds existed that sort of madde it workable--but none really made it a workable experience for me.

Gigastudio(v2) could only address 1gb of physical RAM (documented)--but, in reality, it loaded after tweaks about 720mb for ME. When I asked on different forums AND at Sweetwater, whether the VSL Chamber Strings would be able to load enough on one GigaPC to do a quartet arrangement, I was told variations on "it depends on how you work". The real answer was "not unless you load one section, sequence, and print it as audio before loading the next."

Using a PC VSTi host to host the VSL VIs, while sequencing from my Mac. Unlike Gigastudio, where you load them and forget them...the VI's assume you can interact more regularly with their GUIs...making that not a wonderful solution. Maybe with a VGA switcher and two sets of mice/keyboards...

My "String sequencer box" and it's budget. I had a GigaPC. I had some solo string samples, but wanted small section samples. So, I decided I would bump the RAM to beyond what Giga could address and buy the VSL Chamber Strings. $800 (samples+RAM). Unusable. Non refundable. If I got the plug in versions, I would need a host, and could no longer run my solo samples in Giga at the same time. After much pushing, and thanks to VSL themselves fighting for me, I was able to exchange the ChamberStr for Giga for the plug in version. At the same time, I bought the solo strings from VSL (add $500). Needed a host...after some experimenting with standalones (see above), I decided they needed to live in a sequencer. Bought the competitive upgrade for Cubase4 (thanks to my elderly copy of DP3)--add $400. Since I invested that much in software, I thought I should grab and ADAT IO audio interface while they still made PCI-Echo Gina3G-add $300. I'm borderline with the amount of RAM--I max it (1.5GB) easily and have to be very careful what I load per project. So, at the least, I'll fork over another $200 to go to 3 or 4 gb. Throw the 3gb switch on Cubase, and I should have the head room I need to not be as careful.

$2100 + PC cost (which for now we won't count)...for a string sequencer. If an "expert" had told me straight up "to do what you need to do you'll have to put more than $2k into the box"...I'd have likely passed. With the computer...$3000 string sequencer box-great sound, that I would use on occassion? Pass. But, doing the bit by bit..."well, it will take this little more to get it to work"...has gotten me there anyway.

It's not mysterious stuff. I'm sure there are folks in the world who knew this. But, on one hand, you have forums of users with limited experience but loud mouths...and paid consultants who make their money on selling you things.

In a market that's so "non returnable" as software...and even many computer components that are "exchange only"...shouldn't people have some guidance? Isn't that worth some money?

This got put into my head on another forum today. Someone was "advised" to buy some PC laptop running Vista and Sonar to do DAW multitracking and MIDI. They spent the money. It doesn't work well. I wanted to post "duh"--it's A)a PC laptop B)an ultra new version of, IMO, second rate software C)running under a new OS with documented issues with audio production...it's not his fault. So, I didn't post. I had nothing constructive to offer. It just pissed me off. Just because I know more than HE did about those components...doesn't make me an expert. See above and my many missteps in the DAW world. And I just highlighted big ones that were 100% predictable (by people who've used the pieces).

There should be somewhere you can go and ask things and get solid answers. Keep that in mind when posting here or anywhere. I'm no expert. And unfortuantely, I know far more than most about them. Anyone know of even a pay service for DAW computer consulting? I know there are some system builders that do some consulting...but, in my case--I wasn't looking for a new system. I would venture MOST people aren't looking for a whole new system when they have questions.

We need an expert.

Slightly related thought--Hell, even though these apps are all super deep, most of us know how to use 5% of them. At work, we send people for week long Excel training. We're still told they're using 20% of the functionality. Where's my semester of "Cubase 101"? "Orchestration using VSL's Vienna Instruments"? I'd pay college tuition for those two classes...you know, now that I've got $3k in a string box.

Damn I'm glad I'm not imaginative enough to get into composing with midi stuff. But I hear you on the ill-advise. I've spent a fair bit of coin not being able to see two steps ahead.

Lance

Don't want to be an ass, but I think I warned you.
1.5Gb is a funny number for VSTi, you can halfway work with drum samples with that memory.

GIGA never worked for me, I haven't seen a single person who was able to use it.

I'm not pro, but I've been using this crap for about 10 years.

quote:

Yondan
Picture of nbarts

Posted 04-16-07 01:40 PM Hide Post
Soft synths are my fave in computer world. I've been a fan of them from the beginning.

I run 3GHz P4 HT with 1.5 GIG, no Apples here.

First 1.5 Memory isn't shit with todays softsynth. I easily run out of it. 4 GIG would get me close, but I would run out of my processor power before I get there. It all depends: depends on how full you want your samples loaded & bufer size, bit/sample rate.

So if we are talking about 24/44...

I've never had a chance to use BFD expansion packs, only the base version 1.0. Had no problem using it on a system identical to mine by ITSELF.

I can tell you for example The Grand from Steinberg standalone with all the features enabled overruns my processor when I use sustain pedal. But it runs fine when I playback midi from Nuendo with buffer size 256. It might be fine if I tried to decrease it too, I just never did.

Now with giga & other soundfont files things can get complicated. I don't use Giga Sampler, every time I've tried to run Giga it crashed on me, with numerous errors even after uninstalling it. I had to perform some big time surgery if not reinstall everything after it, so I gave up on it.
But I constantly use giga & other soundfonts with other sample players, including Kontakt & Virtual Sampler. So everything here depends on how your software is programmed. Some require to load the whole font, others just part of it, most of modern will let the user configure this part. So if you want the best quality you will obviously need to fully load the file. Therefore you'll need as much memory as your file is. With todays average piano font size being about 250 Megs, well .... just do the math. Strings & brass are another pain in the ass - these are memory hungry beasts, most of the time you will need to load many variations of the same instrument & use multiple tracks to get one section together.

So for me personally processor power is more of a concern than memory at the moment.

In other words everything is on entry level real time performance at the moment. On todays top level system I can imagine they will all run good standalone, but I wouldn't expect to run many of them simultaneously.

quote:

Each of those seven parts weights about 1.7 gigabytes.
So if you need a violin section fully loaded, guess how much memory you need Smile . If you wished to have a fully controllable strings section with 2 violins, violas, cellos & basses that will cost you about 8.5 gigabytes of RAM.

quote:

www.dawexpert.com

If you really have time to mess with that I can probably host it on my server, but I guess you need to become a "pro" first

Because I'm usually misguided and constantly headed in the wrong direction I have caved in to the KISS(Keep it simple stupid) system of recording.
No Midi, nothing fancy just straight forward get the tone and record it.

Pop, I admire your ability to stay on the cutting edge.
Knowing what not to do is just as important as knowing what to do. Unfortunately Recording software is changing so fast, and there's so much new stuff it's darn hard to keep up.

I respect the information you provide, as far as the learning curve goes you're way ahead of others, and so far ahead of me I don't even know what your talking about most of the time so I have to Google it, but I'm learning that way.

So anyhow thanks for being here

quote:

So, at the least, I'll fork over another $200 to go to 3 or 4 gb.

I've just upgraded my memory to 4G (2x2gb sticks) & sold my old sticks. Overall 1 to 4 Gb upgrade was 60$, I don't know what kind of memory you buy that you need 200$ for it.

By the way, there is a good chance that your older system won't be able to recognize 4Gb of memory anyway & you will probably run out of your processor power far before you reach 4GB memory load.

My CPU doesn't touch 20%. Sampling is ALL about RAM and HD speed.

I've never had a SINGLE problem with Gigastudio itself. So, you can count that as one person you know. It was designed to replace hardware samplers-convert a relatively cheap PC into a disk streaming sampler. They made a revolutionary product. It's conflict with Cubase, I might add, at the time, Steiny claimed was Giga's fault...Giga claimed was Steiny's fault...and multiple folks seemed to think it was Maudio's drivers. Fast foward a year...same driver revision but for new OS (XP). Same Giga. New CubaseSX. Problem gone. So, I'm pretty confident who's fault it was.

RAM pricing...I just went out to Mushkin and priced the same RAM I have now, but 4 1gb sticks. I guess I could sell what's there now to offset. Think it'll bring $30? Motherboard's fine. Ironically, it will address 8gb. But, can't run a 64bit OS...so, I have no clue how you would actually test that. Some flava of Linux maybe--that has Apple's 64bit memory addressing without being fully 64bit. I always thought that was a neat trick of theirs.

quote:

No Midi, nothing fancy just straight forward get the tone and record it.

When I can...I'm with you. You know I do my audio production on an Akai DPS24-and outboard comps and FX...80% of my keyboards are hardware-either digital or my beloved C3/122. I'm a firm believer in the KISS. This string thing...well, I've always LOVED nice string arrangements in pop...and was blown away by how truly real they can be done now. Still, I don't think I'd have put the money into it that I have if that first $800 or whatever were returnable.

MIDI is SO not cutting edge. In fact, it's a pretty solid statement to say it worked FAR better and easier before computers had anything to do with audio. I'm amazed at the relatively lame MIDI tools this current generation have. Granted, having been sequencing for 15+ years, I'm sure I take some of that knowledge for granted, but...the only thing that's improved in the last say 10 is the ability of apps to draw out scores well in real time.

quote:

If you really have time to mess with that I can probably host it on my server, but I guess you need to become a "pro" first

All joking aside, I've considered it...meaning...get a bunch of computers in here...software out the ying yang...Apple, PC, ADK-all kinds. Old ones, new ones. I could charge to test stuff out. Have clean installs of them all imaged...wipe it, install whatever software they're asking about--make some noise. See what kind of issues, if any pop up.

Story about the small windows of knowledge we all have. Local composer did some strings for me. To make it easy on him, I made him a CPR file with everything in the right place in time. He couldn't get his Nuendo to open it, so he imported the WAVs...but, didn't bother to right click them and select "move to origin", which in Steiny's world moves them to the timestamp embedded in the BWAV files. So, he did all the strings completely out of time to my project. Point being-of all the work he's done, including producing some semi major label records, he'd never collaborated or moved between systems enough to even know WAVs could be timestamped--or how to treat them when they were. Since I move between the Akai and PC, if I'm an "expert" at anything, it's in sync and transfers and timestamping.

quote:

So anyhow thanks for being here

No prob.

Someone read me a quote today about "wisdom is the cumulative understanding of what we don't know". Ha. Maybe that makes me wise, after all!!

quote:

My CPU doesn't touch 20%. Sampling is ALL about RAM and HD speed.

I don't really agree, Mr. Popmann, maybe years ago it was. I don't know your sample library and player, but most of the new stuff goes through convolution processing, which is very processor hungry. Add a few more tracks with the rest of your project and there you have it.

And there is a little bit more than your motherboard supporting it, do you know if your OS does? How about your older DAW or whatever software you are using?

they're not really smaplers at that point...but, like NI's Akoustic Piano was like that. But, since it also didn't sound all that great...what kind of instruments are you using that use convolution processing? Just curious.

The VSL stuff is straight sample playback--all the CPU has to do is manage the "what transition sample to play NOW"...My Giga maxes polyphony at 96 voices, which I never have a problem hitting...

quote:

And there is a little bit more than your motherboard supporting it, do you know if your OS does? How about your older DAW or whatever software you are using?

yeah...do some searches for the XP "3gb switch"...allows large memory addressing, which you can allocate how much goes to user mode and how much to kernel mode. C4 supports it officially. Vienna Ensemble supports it. From what I've read, in reality, it will let C4/VSL address about 3.2gb if you have 4...something like 2.5 if you have 3. But, I'll have to juggle that some if I still want to run Giga, which runs in kernel mode. As long as it can grab a few hundred mb, I'll get what I need from it.

Hi Popman
I've been out of the loop for while but I remember you helping me with some computer issues before I got my Mac. Thanks again

I just want to send you my condolences for your problems. I understand. It seems we are in a wolrld where things have become so complex that consumers are now the final reserch and development "drips under pressure". I don't know how many times I wish I could curl up next to a real pro who is in the in the main flow of things and up with the latest and greatest so I can suck up all they have to offer. But unfortunatly I end up paying that expencive tuition for learning by buying gear that in too short of a time is obsolete or I out grow. I'm in the process now reserching preamps, mics, and A/D converters so I can upgrade my front end. I have champagne ears and a Ripple budget. I'm also in the market for a atraight job to pay for it. You may have heard this one but it applies...What do you do when a musician shows up at your door?......Pay for the pizza.

best of luck

Kent

At this moment I don't use anything but drums. I'm pretty sure when I get the right drumkit I won't be using them too. Strings I use (very not often) are giga soundfonts, but I still need to run a convolution verb on top of them to get the feel. Once in a while I fire up the grand, most of the time for solo piano compositions. Other than this all have have in softsynth category are my old SF2 & GIGA libraries & honestly I'm not planning to expand, there are better places for the money to be spent & I'm already way over the line anyway.
These things are good for radio guys for making jingles & stuff like that, this is the reason I used to use them a lot.

Still - drums softs are using convolution, the grand from steinberg, akoustik piano, stuff from gurritan (is that how you spell it?) seems to use convolution, I've noticed that pretty much everything that has a dedicated VSTi player has built-in convolution.

Don't confuse Soundfonts and Giga...FWIW...opposite extremes of size. I have both--actually own some SF drums I LOVE...that's about it, though. Couple 80's digital patches that were sampled in SF.

Anyway--I hear you on the convolution. Akoustik Piano was an awful hog. You could turn off that ambience...it certainly played better, but it left it clear how nasally the samples are.

I'm pretty sure the VSL stuff is just playback...it DOES have to figure out which transition samples to play in real time, though. We'll see how that goes once I have 3gb of RAM buffered up and notes flying around...

Is BFD? What's is drum software using convolution for? That makes no sense. The entire POINT of those libs is that they sample every hit through all the mics, for blending later...