HD Native vs HD6 feature mixing shoot out

[JSt0rm]

Originally Posted by UnderTow

Same here.



I'm not telling anyone what to do. :-) I'm just curious if there might be some aspects of ReVibe I am not aware of. I tend to use ReVibe myself a lot for post simply because most studios have it and I quite like it's sound.

Btw, is anyone using the Sonnox Oxford Reverb for post? It exists both as an RTAS and a TDM plugin, hence the question. (There are more RTAS only Reverb options).

Alistair

Yeah sorry I didnt mean you directly. I think its easier to automate something like revibe and when you have deadlines and clients breathing down your back its best to go tried and true.

[seancostello]

Originally Posted by kdm

I would look into ValhallaRoom. Excellent reverb that can actually work well as either a room sim or traditional effect.

Uses a little cpu, but lower than Aether or convolution.

Not surround capable though. I need to talk to the developer about that.

Talk to me. Feel free to ask about this in the ValhallaRoom thread in the New Products / Products older than 2 months subforum - but I would welcome any thoughts/requests for surround reverbs.

EDIT: I set up a thread for feedback here:

Reverb Designer Seeking Input Into Surround Reverb Plugins

[AMIEL]

From what I know, but somebody correct me if I am wrong...

Using any PT HD TDM System and only RTAS Plug ins..... is not the equivalent of having a HD Native ( but the number of voices dependent of which TDM System u have) plus the same plug ins and even more??? because in HD Native part of the power of your computer is used to play the tracks??????

For Example Using a Mac 8 Core with a TDM System like a PT HD3 Accell gives you a voice limit of 192 tracks (256 total) at 44.1/48 kHz ....plus all your computer power for Plug ins???

In my case if this is true, I am happy with that track count plug all that power for RTAS Plug Ins.....plus I can run some TDM plug ins that do not exist in RTAS format...

[JSt0rm]

Originally Posted by seancostello

Talk to me. Feel free to ask about this in the ValhallaRoom thread in the New Products / Products older than 2 months subforum - but I would welcome any thoughts/requests for surround reverbs.

EDIT: I set up a thread for feedback here:

Reverb Designer Seeking Input Into Surround Reverb Plugins

Make it. The thing is it need to be worked with quickly and have nice automation. A mixer needs to be able to adjust some perameter quickly for a section of a project without making a big ordeal.

"I like it but make the reverb bigger"

"you want it louder?"

"no, just make it bigger without being louder at all and make it run into the next scene but only at the end of the scene on the last shot"

With revive the mixer can just crank the tail highlight the section and snap it for that bit of time.

It's just faster to get to the same place. Also revibe looks like teh sexhy time and a lot of mixers leave a few choice plugs open on the second display all the time

[sound_music]

some great suggestions for ReVibe replacements, cheers.

it's going to be a hard one for me to replace, definitely the TDM-only plug i'll miss the most. i use it on everything...

[BIGBANGBUZZ]

Yep.... Been saying this a while now TDM =dead

Sent from my GT-P1000T using Gearslutz.com App

[BIGBANGBUZZ]

Originally Posted by G-Spot

The number of i/o's and TDM plugins only like HEAT...

Heat pretty useless for post?

Sent from my GT-P1000T using Gearslutz.com App

[brandoncross]

I just want to point out the processing difference between TDM and Native. I recently upgraded my DAW to a w3670 xeon processor (600 usd) and ran a quick D-verb test with my HD Native card. I got 620 D-verbs! Things got a little laggy at 620 so I stopped but I think I could have gotten 20 more instances. An Accel card can open 35 d-verbs. A little quick math and the 600 dollar processor is equal to an hd 17 accel! Bottom line is it doesn't matter how cheap an accel card is, native is just much much better. By a great margin.

[renec]

Gary,

It'd be cool if you could do a little more extensive testing when you have an opportunity. spending a moment to really find the breaking point of a native system could yield very useful results for the community at large.

also, it'd be interesting for anyone mixing high track count shorts to attempt one on a native rig. I'm sure that there are other caveats wrt film mixing in native that have not yet been sussed out.

[UnderTow]

Originally Posted by brandoncross

I just want to point out the processing difference between TDM and Native. I recently upgraded my DAW to a w3670 xeon processor (600 usd) and ran a quick D-verb test with my HD Native card. I got 620 D-verbs! Things got a little laggy at 620 so I stopped but I think I could have gotten 20 more instances. An Accel card can open 35 d-verbs. A little quick math and the 600 dollar processor is equal to an hd 17 accel! Bottom line is it doesn't matter how cheap an accel card is, native is just much much better. By a great margin.

Just to be sure the comparison is fair, at what latency did you run the HD¦Native project?

Alistair

[Kuba_Pietrzak]

From my point of view, the most interesting thing is, how a native system responses to a fast and intensive user's gestures, which is fast editing, programing automation, quick starts and stops, scrub here and there and navigation within the session.

We all know, that native systems can handle huge post projects - there are Nuendo users among us - and we heard many things about doing large projects on this native platform.

But question is, how such a HD|Native system really behaves during an intensive editing/mixing session.

Really interesting...

regards,
Kuba

[brandoncross]

64 sample hardware buffer

[Airon]

Ouch. We've switched to HD native as well. I'm running my home rig with my RME Fireface UC at 128 samples with no trouble.

Reverb recommendations are the same as have been mentioned. 2C Audio Aether and Valhalla Room. The Lexicons are pricey, familiar and efficient. All can be tested.

Nice to hear that Altiverb 7 is on the way. Almost thought they'd all gone on holiday to record more impulses or something .

[nst7]

For those of you wondering about selling your TDM systems, vs. keeping them for a potential tradeup to a new TDM systems if it comes out, the AES conference is happening in about 2 1/2 weeks, which is generally when Avid launches new products.

If it's this close, I would encourage you to at least wait and see if something new comes out before you sell anything.

[ggegan]

I don't have any plans to do anything prior to AES. I am also waiting to see whether Apple announces a new Sandy Bridge Mac Pro with a Thunderbolt port.

[sound_music]

don't hold your breath for an announcement from AVID @ AES, we've been hearing that exact rumour for the last 4 or 5 AES shows!

(although it would be great if it was actually true this time ... i'd love to be wrong about this )

[ggegan]

Getting back to the topic of RTAS reverbs, I just downloaded the Flux Verb demo, and I have to say that it sounds excellent, I think even better than Revibe. The GUI is a bit dark and I couldn't figure out how well it automates, but the quality of the reverb is very realistic to my ear, especially for post work, and not splashy unless you want it to be. Anybody have any first hand experience with it?

[Brent_in_Sydney]

This thread inspired me to do a similar shoot out between my HD3Accel and newly installed HDNative card on PT10. Thought some of you might be interested:

Brent's HD3 vs HDNative test

[BIGBANGBUZZ]

Originally Posted by kdm

I would look into ValhallaRoom. Excellent reverb that can actually work well as either a room sim or traditional effect.

Uses a little cpu, but lower than Aether or convolution.

Not surround capable though. I need to talk to the developer about that.

Best reverb I've ever owned

[jumpnyc]

Originally Posted by Brent_in_Sydney

This thread inspired me to do a similar shoot out between my HD3Accel and newly installed HDNative card on PT10. Thought some of you might be interested:

Brent's HD3 vs HDNative test

That's amazing Brent!

[NReichman]

This thread inspired me to do a similar shoot out between my HD3Accel and newly installed HDNative card on PT10. Thought some of you might be interested:

Brent's HD3 vs HDNative test

Back when the Complete Production Toolkit first came out (PT8), I was raving about how powerful and fast the all-native option was. Most HD users didn't take me seriously. So, 2 major versions of Pro Tools later, I'm happy to see Brent talking about the same test I performed back in 2009.

[Brent_in_Sydney]

True - but back then it didnt support ICON or destructive punch or lock up to tri level sync for laybacks locked to LTC etc. CPTK and 8 were great but HD Native and 10HD sw are truly beastly on first impressions - especially when you add in 16k of ADC. I'll know more once my next project starts in earnest in 4 weeks how it performs but Im flabbergasted how the same CPU on the same settings with the same session can apparently absorb the power of a HD3 through seemingly 64bit optimisation vs a 48bit fixed mixer.
TBH (and no disrespect to earlier comparisons) I always suspected the native converts were comparing TDM only usage vs native not (48bit TDM+Native power) vs (64bit Native), that's why I wanted to test it myself - details on the previous threads were always a little bit thin on the ground and I'm just a bit overly detailed at times (OCD?) - lets face it, every system is a bit difference, maintained differently and its not like Avid are going to post details real world comparisons any time soon.

[ggegan]

Yes, I was comparing PT9 HD6 TDM only usage to PT9 HD Native usage, which may be a bit misleading because TDM systems can use both. These days I have been populating all my audio tracks with RTAS plugins and using TDM for all the back end processing, ie reverbs, 5.1 bus limiters, surround mix engines and delay compensation. On many feature projects I easily max out both types of processing.

I am running the HD6 on an 8 core Harpertown, which is quite a bit less powerful than the 12 core Westmere I used to test the HD Native system, so the HD Native was very competitive. I'd say it was more powerful than the TDM HD6 only, but definitely less powerful than HD6 TDM plus the Harpertown native processing, but not by as much as I had expected. I was duly impressed.

[tom_lowe]

People are finally realising native is better in most cases and at very least as good as a TDM system, this is why I can't believe Avid are still trying to sell new hardware. If a 12-core Mac currently beats and HD6, then surely the Ivy Bridge CPUs (maybe in a Mac, most likely not) will beat the HDX that Avid say is equivalent to an HD10. Since you need a fairly recent Mac Pro to run HDX (would make more sense if you could use it to upgrade a Mac Pro 1,1 etc) then isn't this utterly pointless, since the more power PC would be far cheaper.

[ggegan]

Originally Posted by tom_lowe

People are finally realising native is better in most cases and at very least as good as a TDM system, this is why I can't believe Avid are still trying to sell new hardware. If a 12-core Mac currently beats and HD6, then surely the Ivy Bridge CPUs (maybe in a Mac, most likely not) will beat the HDX that Avid say is equivalent to an HD10. Since you need a fairly recent Mac Pro to run HDX (would make more sense if you could use it to upgrade a Mac Pro 1,1 etc) then isn't this utterly pointless, since the more power PC would be far cheaper.

The HDX hardware makes perfect sense for feature film mixing where you need massive numbers of voices and don't want to link too many multiple machines. An HDX2 provides 512 voices and supposedly the equivalent of an HD10 TDM plus whatever native processing is available. With an HDX3 that increases to 768 voices and a third more card processing.

While even an HDX2 is going to be overkill for the vast majority of users, for those mixing Hollywood features it is cheaper and makes more sense to go with one HDX2 or HDX3 than the equivalent number of HD Native systems that would be required to provide the same number of voices and needed processing. Remember that each Native system will need it's own powerful computer and a Sync HD as well as duplicate licenses for any non bundled plugins. That is a lot of extra expense, nevermind the hassle of dealing with mutiple systems.

The issue of high end native algorithmic reverb plugins is also still a question. There will reportedly be native versions of Revibe and Reverb One coming out, but no one knows how much CPU power they will gobble up.

[philper]

So....will dubstages with big TDM investments go native...soon? Never? Only when the old gear wears out? When a software update forces them to?

phil p

[KMcK]

Originally Posted by philper

So....will dubstages with big TDM investments go native...soon? Never? Only when the old gear wears out? When a software update forces them to?

phil p

They already are. There are 8 brand new stages in the building below me, using HD Native for all the playback machines (6 satellites per room). TDM hardware is only being used on the stem recorder machines.

[tom_lowe]

Originally Posted by ggegan

The HDX hardware makes perfect sense for feature film mixing where you need massive numbers of voices and don't want to link too many multiple machines. An HDX2 provides 512 voices and supposedly the equivalent of an HD10 TDM plus whatever native processing is available. With an HDX3 that increases to 768 voices and a third more card processing.

While even an HDX2 is going to be overkill for the vast majority of users, for those mixing Hollywood features it is cheaper and makes more sense to go with one HDX2 or HDX3 than the equivalent number of HD Native systems that would be required to provide the same number of voices and needed processing. Remember that each Native system will need it's own powerful computer and a Sync HD as well as duplicate licenses for any non bundled plugins. That is a lot of extra expense, nevermind the hassle of dealing with mutiple systems.

The issue of high end native algorithmic reverb plugins is also still a question. There will reportedly be native versions of Revibe and Reverb One coming out, but no one knows how much CPU power they will gobble up.

Yes, for the time being they do make sense. However, the point I was making was that considering HD6 is already outperformed, it won't be long until an HDX3 is outperformed. CPUs improve all the time, it's taken Avid 8 years to upgrade their proprietary hardware. The performance/cost ratio is also much lower when not paying the Avid premium.

GPUs greatly outperform HDX cards in terms of bang for buck, Avid and other companies should utilise these for audio processing instead of making proprietary, locked-in systems. In terms of sheer power, an nVidia Quadro 6000 costs around £3,500 and has a far superior spec to that of an HDX card. If Avid were to utilise OpenCL and/or CUDA, they'd be on to a winner. I/O dealt with by Thunderbolt, audio engine using a multi-core Intel i7 and nVidia GPU.

[awalk]

Originally Posted by tom_lowe

Avid and other companies should utilise these for audio processing instead of making proprietary, locked-in systems. In terms of sheer power, an nVidia Quadro 6000 costs around £3,500 and has a far superior spec to that of an HDX card. If Avid were to utilise OpenCL and/or CUDA, they'd be on to a winner..

I know Merging Technologies has gone this route with amazing results.

[tom_lowe]

Originally Posted by awalk

I know Merging Technologies has gone this route with amazing results.

Oh when did they do that? I looked in to them a few years ago and they were still trying to flog the Mykerinos card. If I'd known they were doing something sensible, I may have been persuaded!