64-bit Processing Architecture

[soybalm]

The VI's and everything else will benefit by allocating CPU resources due to being 64 bit. It's should be way more organized. A task like file linking won't bring Pro Tools to its knees like it does now. There are lots of benefits if and only if the programmers utilize them. According to Avid, they wanted to rewrite the program to make sure they were making the most out of what 64 bit programming has to offer. Some programs can just become 64 bit and boot up but will probably run like crap. It took so long for IE 64 bit to run well and it still has problems. I think Avid did there homework. Pro Tools Expert has a great added video speaking about how ProTools doesn't fall over when throw all kinds of tasks at it. We'll see.

[Eric Lambert]

Originally Posted by Sardi

Sure. But as I tried to explain, the ones benefiting from the extra RAM availability, are instruments loading large sample libraries.

I'm not following your point. Those are precisely the plugins which NEED extra RAM. Ivory, Omnisphere, SuperiorDrummer... these are session killers. They force my sessions into the RAM ceiling in a matter of a few minutes. Opening the roof for them removes one of the most frustrating things about ProTools I've ever had to deal with: limiting my choice of instrumentation for the mere sake of RAM.

Access to more RAM is huge for composers. HUGE! I can't say it enough.

[sunburst79]

Both types of VIs will benefit. Sample based more than processor based but even a CPU bound VI will benefit from the new AAE.

[Sardi]

Originally Posted by Eric Lambert

I'm not following your point. Those are precisely the plugins which NEED extra RAM. Ivory, Omnisphere, SuperiorDrummer... these are session killers. They force my sessions into the RAM ceiling in a matter of a few minutes. Opening the roof for them removes one of the most frustrating things about ProTools I've ever had to deal with: limiting my choice of instrumentation for the mere sake of RAM.

Access to more RAM is huge for composers. HUGE! I can't say it enough.

The title of the thread is 64-bit Processing Architecture.

I pointed out the following.

Originally Posted by Sardi

This is not a personal dig at you, but I wish people would stop saying this. That analogy only applies to VI's that use huge sample libraries like Kontakt instruments.

The fact that PT has always been inefficient with all types of VI's won't magically disappear by having access to more RAM. Most soft synths don't require large portions of RAM at all.

What will (hopefully) benefit VI users is the fact that the audio engine and plugin architecture in PT11 has been re-written from scratch. This is where we are going to (hopefully) start seeing vast VI improvements.

Then this whole back and forth with people that don't seem to grasp the concept of RAM is great for sample based VI's but doesn't explain the fact that PT ran like crud with standard synth VI's that don't require bulk RAM led me to say (again) this.

Originally Posted by Sardi

Sure. But as I tried to explain, the ones benefiting from the extra RAM availability, are instruments loading large sample libraries.

Listen carefully to the VI's he talks about. They're all large library plugins like Structure, Kontakt etc. From this point on you can here him say it:

http://youtu.be/W0dmMaLsAuY?t=56s

What I am trying to explain is that plugins like Massive, FM7, Hybrid etc. won't necessarily benefit from having access to more RAM as they're not loading libraries. They are synths using oscillators. These types of plugins never really ran efficiently in PT10 and earlier, especially compared to other apps like Logic etc. even when those apps were 32 bit.

So again, it's the new plugin architecture and engine that is going to (hopefully) dramatically improve general VI usage in PT11, not RAM. RAM will be integral to loading large library VI's.

That's what I'm trying to get across. It wasn't lack of RAM that caused PT to be inefficient with VI's. This only applied to large sample library VI's.

So, I don't see what it is that there is not to understand about my point. I think I've made myself perfectly clear... repeatedly.

Either people have reading comprehension problems or they just don't want to understand the difference between a synth VI that might only need 50Mb of RAM access compared to a sample based library VI that could possibly need gigs of RAM access.

All I've been trying to say, is that from a VI point of view, the new engine and new plugin format is where we are going to say MASSIVE improvements on how many we can run simultaneously. RAM access will play a BIG part for sample library based VI's, but that was not the only bottleneck prior to PT11 for instruments. Running a handful of Hybrids or Massives used to make PT choke. That had NOTHING, I repeat NOTHING to do with RAM access.

How the hell do you explain the fact that Logic was able to run multiple times the amount of instruments that PT could run even before it was ported to a 64bit app? Or the fact that even if you ran Kontakt using it's own memory server, it still wasn't efficient in PT compared to other DAW's.

I feel like I'm just repeating myself constantly here. Echo Echo Echo....

Sorry if this is coming of a tad harsh. I don't mean to offend anyone. It's just frustrating when people think the only benefit to PT going to 64bit is the fact that it can access more RAM.

[Eric Lambert]

Originally Posted by Sardi

It's just frustrating when people think the only benefit to PT going to 64bit is the fact that it can access more RAM.

I think you're frustrated, unnecessarily, because you're repeating points which aren't gaining attention precisely because it's unnecessary to make them. There are MANY advantages to access to additional RAM - we all agree - but your previous point (different from the one you're making now) was that primarily sample-based VIs will benefit. We've been disagreeing. Now that you've changed your mind we are all in agreement.

PT 10 and below were problematic for many reasons, not solely, or even primarily, because of the 32-bit architecture.

[Sardi]

Originally Posted by Eric Lambert

Now that you've changed your mind we are all in agreement.

Last post as this is just getting ridiculous now.

Where exactly did I change my mind?

I've re-read this thread 3 times now and either you've mistaken another post for mine or I must be going crazy as all I've done is say the same thing over and over.

[nigelpry]

Sardi .... you haven't changed your mind. The point you made was well understood, by me at least, from the start.

I'm a heavy user of both types of VI, and I'm hoping that PT11 will see big improvements in VI efficiency.

Yes, access to extra RAM will supposedly mean we don't need to use Kontakt's memory server any more, although I'm waiting to see it in action before I come to a firm conclusion as to whether that's the case.

I'm hoping PT will manage my VI RAM needs effectively, rather than me having to reserve a portion of RAM exclusively from Kontakt. I use a few other memory hungry VI too.

But I'm also hoping that my Arturia and NI synth plugins will benefit from efficiency gains too.

Of course I'm crying about my G-Force plugins .... It's been confirmed that there are no plans to port any of the existing plugins to AAX in the forseeable future ;-)

impOSCar (not sample based) and both M-Tron Pro and Virtual String Machine (sample based) are among my 'go to' favourites, usually found something to do on most tracks that I can get away with using them on.

[Sardi]

Nigel, thank you.

You seem to have grasped EXACTLY what I have been trying to say. Obviously you have faced the same issues most of us have with VI usage in PT, and understand how much this new engine and plugin architecture will help us.

[Eric Lambert]

Originally Posted by nigelpry

Arturia and NI synth plugins will benefit from efficiency gains too.

Arturia VIs are beasts. The new engine will open the door for better performance. Whether that's actually realized in PT11 is another story -- we'll have to wait and see.

Originally Posted by nigelpry

Of course I'm crying about my G-Force plugins .... It's been confirmed that there are no plans to port any of the existing plugins to AAX in the forseeable future ;-)

I've heard this, too. It will be a shame to be without M-Tron, which is in heavy rotation at our studios. I suspect that they'll eventually port to AAX but it won't be soon. Certainly not soon enough.

[ThunderKyss]

Originally Posted by Sardi

Exactly. Hopefully now we will have a powerful and amazing performing piece of software. With a new engine and plugin architecture, the issues we faced before should hopefully be a thing of the past. :)

Originally Posted by Eric Lambert

PT 10 and below were problematic for many reasons, not solely, or even primarily, because of the 32-bit architecture.

If it was(is) so bad, why do so many people stick with it? Why isn't some other more efficient, better written program the go to application for audio production?