If it has Ethernet ports on the back, that's Dante you are looking at. The Dante card is for performance. You can use these units without a Dante card, using your standard NIC card with an emulator driver, how ever the performance would be significantly decreased. You can also plug these units into a switch. While the Ethernet ports are also there to work with the existing rednet system just the same as the digilinks are there to work with existing protools HD interfaces.

I'm actually waiting on Part 2&3 of the review by Pro Tools Expert for clarification on the Dante functionality. I dont see why you can't connect it with a computer via Dante. The only thing is that you might have to map the I/O from the interface first. For example to send Mic Pre Ch 1 to Dante out Channel 1 from the interface and so forth.
With regards to Pro Tools HD it works well with HDX and HD Native.

Hopefully the guys at Pro Tools Expert don't keep us waiting for too long.

Here is Part 1 of the Review.

Very cool... Anxious to hear parts 2 & 3.

This is a key question for HDX/HDN users. The latency numbers on the Focusrite site do not match any HD I/O numbers I have measured. With this in mind I worry that this interface will not be sample accurate with hardware inserts and will not place audio on the PT timeline correctly (it HAS to mimic a n AVID to do this correctly). Apogee and Lynx have always gone to great lengths to make their I/Os match AVIDs. Thats why they are sample accurate. My fingers are crossed hoping that the Red4 Pre has done this as well. If not - the Apogee ensemble 2 (edit - I meant to suggest the Apollo Symphony II here, not the Ensemble II) would make the much better choice (but you need to add TB and DigiLink cards...

Can anyone answer the tech facts about sample accuracy in PT using Red4 Pre with HD/HDX/HDN.

If this interface can match the AVB for sound quality, I will be very impressed as the AVB sounds incredible. My previous experience with several Focusrite Interfaces is anything but favorable in regards to sound quality, however, I would be happy to be impressed as this does look like a good product. I just don't like when they use reverse marketing to be elusive as follows....

"From the Focusrite website....

"About the chips
Strange to say, if you’re dealing with a reputable manufacturer, knowing what converter chips they used is fairly immaterial. You should be able to expect that they will have chosen the best part for the job – and that will include knowing that it won’t become obsolete during the expected life of the product as well as performance. And of course price will come into it too. In the old days, when some chips gave markedly better performance than others, some manufacturers erased the identification on their converter chips so nobody else would know what they using. Today… there’s really no point.

Chips alone are not confirmation of conversation quality, however, hiding the information under marketing avoidance is not impressive either. Additionally, I am unhappy with Focusrite continuing to live off the legacy of the ISA and Red range. If they truly believe in this interface, they should package it with it's own skin, especially with the set of features it has. Leave the Red range legacy out of it. The pres in this interface do not bare any resemblance to the transformer coupled preamps of the original Red 1 and Red 8 preamps, so other than the obvious marketing benefits of legacy association, there is nothing about the original red range included within this interface, well, accept for the red military grade aluminium face plate

I find that hard to believe as this unit is actually a derivative of the Rednet with better conversion as its a Dante Ethernet Audio Interface with Thunderbolt. This unit shouldn't be compared to an Apogee Ensemble or Apollo units as its an entirely different ball game. It cost far more than the Clarett range. The Clarett units should be compared with the Apollos, Motu AVB, Apogee Ensemble and UA Apollos.

They are one of the least companies that hide anything about conversion as they don't post manufacture data specs about their conversion performance. They go by the AES17 standards when documenting real world conversion performance A Weighing (db). There's lot of companies out there that like to emphasis chips and post the data chip specs from the manufacture that can be misleading to customers. Motu doesn't even hardly post anything about of half of their interfaces A Weighing Dynamic Range as we really don't know if they are posting chip performance or real performance. They failed to publish the Motu 16 Track Dynamic range as I had to email them to get that info and all they told me was they don't publish those specs but told me that its 112 (db) in and 112 (db) out. I ask them about the ADC chip for the 16A, they didn't want to tell me nothing. You also gotta look at the noise floor, Harmonic Distortion. But nothing on paper can really tell you anything about how something sounds. The best tools you can use, is your ears!

My mistake - I meant the Apogee Symphony II not the Ensemble :-) as a sample accurate HD interface.

Does anyone know the matrix capabilities of this unit? IE, can the dante inputs be directly routed to Analogue and ADAT outputs? and visa versa. So it can operate as a standalone AD/DA onto a dante network, as well as its obvious use as an interface.

Looking at the focusrite control PDF for the clarett series it looks like only DAW Playback and Custom Mixes can be routed to outputs - seems very limiting.

Even my old firestudio 2626 could route physical inputs directly to physical outputs.

If this was cleared up for me i would purchase one today!

Also, i use a rednet PCI card for waves multirack with my console for live gigs.

It looks like this interface will be able to do the same job at the same latency and allow me to do it on my laptop for smaller gigs - albeit only 32 channels.

Just received my Red 4Pre today and tested loopback for accuracy on the timeline with HDX as the host. I define loopback testing as outputting a signal from ProTools and looping the output of that back into an input and recording the result. With all AVID interfaces in HDN and HDX the newly recorded audio lines up exactly (no loopback error).

I have only just started with this unit (it is a beautiful looking piece). As ProTools expert points out (thank you!) the way it shows up in Hardware setup is as 2 Digital AVID I/Os per DigiLink port used. A bit weird and confusing since none of the listed inputs on those pages are right and all Input setup is done on the Red 4PRe itself (like selecting what is in A/D inputs 1-8). Regardless - there are charts in the manual but it is pretty odd - like Main output 1/2 on the DB25 is really Outputs 7/8 from the DAW. Will see after working a bit how to better describe this...

SO- loopback testing. In this first round I found that at 96kHz the loopback error is 35 Samples late and at 44.1kHz it is 30 samples late. Note that in ProTools there is no simple automatic setting to fix this. You either have to move it manually, use DDMF MetaPlug to adjust it (a native only plugin) or come up with a clever way to shift these with user delay comp settings... I will double check all this tomorrow and have already written Focusrite about it. For HDX users especially - we want the unit to be seamless. So far - this is not...

Also a quick check of the minimum RTL at 96kHz with no submasters and no plugins is 0.58ms. With an HD I/O it is 0.48ms. To my thinking - to be a sample accurate clone the clone unit needs to be exactly as fast or faster than a real AVID I/O. If faster one can always add some delay in the unit to make it line up with an AVID I/O. If slower I can't see how one can fix it. As always - everything is measured externally with scope and function generator.

Now all the how it sounds stuff... give me a few days :-)....

Yeah this is the perennial problem with any interface that connects to PT HD - It HAS to have a lower latency than the HD IO in order for the manufacturer to then be able to introduce a sample delay to allow matching to the Avid. As rightly pointed out, if this isn't the case, audio is placed incorrectly on the timeline, and Pro Tools does not allow you to adjust for this. In this sort of scenario you might end up a handful of samples out, or quite a bit, depending on the specific product, and obviously anyone running a third party converter into an Avid digital IO will likely see the most discrepancy as only the latency of the digital transfer is compensated.

It can sometimes be a problem with interfaces that run over Core Audio or Asio into Pro Tools (normal or HD), if the latency reporting is incorrect. If the interface has software that allows a sample delay to be applied, that can be a get out of jail free card provided, again, the latency is lower than what's being reported and thus compensated by Pro Tools.

It's frustrating and...you know...pathetic...that there's no option to manually correct this in the software. I don't have a problem with the sometimes limited nature of Pro Tools functionality, the lack of customisation etc. I get that it's a product which maybe benefits from being 'uniform' and elegant in function rather than extensive. But to deliberately omit this sort of functionality is, ironically, not professional.

There is a workaround for the recording compensation thing, as long as you're recording a single track...You set up a hardware insert, feed the cue mix to the output, and the source to the return, printing it to a new track in realtime, and using the hardware insert manual compensation to adjust it. But how clunky is THAT?

J

Hey Jack - so great to get your response - you totally "get it"! If AVID would just give a field to enter a HW offset (like many DAWs have) this would be a NON-ISSUE!

The bigger issue for me with this kind of error is
Track one is lets say drums
Track 2 plays perfectly to track one but is placed on the timeline wrong
Track 3 now gets two ever so slightly out of sync tracks to play to
Track 3 regardless of what they play to is also placed wrong on the timeline...
Rinse and repeat....

Because Foucusrite chose the routing it did for its I/O map of the 4PRE there is not a single I/O that can be easily used as an I/O insert as you can in Pro Tools. take a stereo pair - PT only can do I/O inserts for inputs and outputs that are mapped to the same number in the I/O window - Like input 1 and 2 are a stereo pair that only work as an insert with Output 1 and 2. In the 4PRE The 8 inputs are numbers 1-8. But the first two outputs are a monitor jack, next four are the two HP jacks and then starting with #7 you get to the DB25 outputs. Analog inserts are possible with INPUT 1/2 and the Monitor Outputs. The IN 3-6 are all mapped to HP outputs -no one wants that for an insert. Then Input 7/8 paired to output 1/2. Then all the Digital ones are skewed according to what sample rate you are at and how many ADAT channels that can do. Would have been SO MUCH better to map Inputs 1-8 across from Outputs 1-8 , then the ADAT ones - then the SPDIF - then all the ones that are Focusrites specific that don't map to an AVID I/O. Or maybe have a different map for ProTools mode....

Regardless - I found at 96K I can hardwire ADAT out 2 to ADAT IN 1 and do a digital insert with an optical cable. Then I set that up to have a HW delay (fictitious) of 1.1ms. Then use this as an I/O insert on the Submaster bus. This corrects the loopback error all the way down to 2 samples. The cost is an extra 0.2ms of RTL (everything you add to the mixer adds RTL). The RTL for using the 4PRE this way and accounting for loopback error is 0.91ms at 96kHz (one Sub master with the I/O insert in the mixer. Using an AVID I/O (no I/O insert needed) you get 0.52ms.

Though I will mercilessly point out what I don't like - the overall grab for all that is good in this unit is still huge (like no fan, and all the connectivity.....)

Next up will be looking into RTL and Loopback for running native TB. And of key importance to me - is 96kHz, 32buffer useable???

Ah that's an interesting solution. I mean, I guess it's not THAT hard to just nudge the audio file after each record pass...but then the time stamp is 'wrong' which might bite you later if things are phase critical in some way. But it's all just silly isn't it? We shouldn't have to. I can track in Reaper and have none of this nonsense (except of course if I want to multitrack because the Reaper 'Takes' system is BANANAS)...Anyway.

J

Logic has a field you just enter the offset - Done
Studio One has a field you just enter the offset - Done
Pro Tools wishes you would only use AVID I/O :-)
10 vocal takes in PT then go shift every one of them... ughhhhh!

Well it's odd - they wish you would only use their IO, but they provide a shed load of digital IO, which will often need to be compensated for whatever is a the other end. And of course they provide a version of PT that uses Core Audio and ASIO drivers, but with no means to adjust for discrepancies that might arise.

J

^^^^^^^^^^^^
You are so right - I always forget about that because I rarely use it....
Truly! All kinds of Digital I/O that leaves out any way to compensate for where its coming from... brilliant ... not...

Back to the 4PRE now :-)

Incidentally it SHOULD be correct using the network...You could run it to HD software/Core Audio Dante Driver...it will rely on correct reporting from the Dante driver of course, and the latency will only be low if you use the PCI card. If you're using DVS it's worth checking to make sure that the compensation is correct - when Audinate developed DVS they got the latency reporting correct for the PC version, but they didn't initially bother with it in the mac. Apparently the devs themselves could not be easily persuaded that it was a priority. That was perhaps a year or 18 months ago. I suspect they'll have done it now, but the PCI card would have correct reporting even if DVS does not on mac. For anyone that's wondering about this, basically with Dante, you specify a network latency that is enough to cover the longest latency device you intend to connect to the network...there are fixed intervals, like 3ms, 6ms etc...can't remember exactly what they are. But anyway, that will then be the latency as long as whatever you connect is less...the network queues packets up and ships them out at 3ms or 6 or whatever it is.

J

Can anyone with a unit tell me if you can route inputs directly to outputs within Focusrite Control?

ie can i take ADAT in 1 and using the onboard matrix send it to Dante Out 1

Like i asked earlier in this thread.

Thanks in advance

** EDIT - I didn't use the function but apparently when using the TB connection you can make at least some alterations to things using Focusrite Control. In ProTools mode this SW interface is not available. ****

I am very novice with this interface and Focusrite control and don't use DANTE at all. In the testing I have done with this unit and ProTools HDX and Logic via TB - the Focusrite control app gives no access to the internal matrix router in the 4PRE. What is listed in the manual for each of the sample rates is how the routing works - with no ability to map any input to any output.

Now if you went with DANTE and whatever control panel it has I don't know what is possible but from what I know of Focusrite Control so far the answer to your question is "no".

Thanks, the dante controller only controls what goes to the Red4 and where the outputs go on the Dante network.

So in focusrite control for an output you can only select Daw playback channels or Custom Mix?

Even my old Firestudio 2626 could route adat channels directly to analogue outputs
:/

You should read the Focusrite Control Manual and see if that helps you... My use of this part of the SW was very limited and I may not really know what it is you are trying to do. Apparently there are ways to route things when not in Pro Tools mode - take a look and see if the manual answers your question.

https://us.focusrite.com/sites/defau...user-guide.pdf

Yep ive read it many times - its not updated for this interface its based on the clarett series. On those interfaces only Playback (DAW) and Custom mixes can be routed to outputs.

if anyone can list what options that are available in drop down boxes for the outputs in Focusrite control that would help.

FWIW - I will be sending my 4PRE back on Monday.

The unit has so many great features but it did not win me over. As reference I have been tracking for months with HDX in a one track at a time mode with acoustic instruments and vocals at 96kHz with select plug ins so that total RTL is between 0.6ms and 0.9ms. IMO and IME (and many will argue I am wrong) the reduction in comb filtering for these super low latencies adds clarity to the sound and much joy to the effort while really letting me hear how things sound as I record. The payoff is that when I play these tracks back - I have very little work to do to finish things. They sounded great whilst recording and pretty much the same on playback... (long discussion for another thread)...

With the 4Pre I expected the HDX side of it to work sonically and not be an issue (and it did). The loopback error is unfortunate but not totally unexpected. The Native performance was the real test. (all of this is 96kHz 32 or 64 buffer, but for a Native interface this is the only area I care about). The things that make me pass on the 4PRE

- Max output is 18dBu where as AVID I/O is +22dBu. Made me run all my monitoring and HP equipment much hotter and especially in HP resulted in more noise (not the 4PREs fault - and this is all just using my normal Outputs feed remote HP amps, not their HP).
- Loopback error in HDX mode
- Odd routing matrix that makes I/O inserts in HDX mode only possible in a few isolated cases.
- 1.7ms latency even at 96kHz/32 buffer (Apogee Ensemble 2 publishes 1.1ms and HDN does this same laetncy at 96/64)).
- Driver efficiency at 96kHz low buffers. The absolute RTL was not a surprise since Focusrite very nicely publishes these but the driver efficiency running Logic was just not good enough for me. Virtually no Reverb I had would even run there and with only 2 tracks, EQs and Comps plus one Space Designer was consuming over 1/2 of my nMP Hex Core. Just not working well enough for me.

The Clarett is a 1k converter and not in the same league as either the Ensemble Thunderbolt II or the Apollo Blackface.

Both the Ensemble Thunderbolt and blackface Apollo are pro units and not prosumer as they easily spec higher than what used to be Apogees(AD16 line) hi end products. Oh and they sound better as well. Lol
Oh and if the Ensemble Thunderbolt or the blackface Apollo was Prosumer they would not be coming in at $2500 per unit. That is hardly a prosumer price.

This Red Pre 4 spec and design wise is aimed right at these products (Apollo blackface and Ensemble Thunderbolt) and not the Symphony MKII.
I would be highly surprised if it performed to the level of the Symphony MKII.

Not saying the Focusrite Rednet converters are not great soundingn converters, but let's get real here. I doubt it's at the level of their top of the line, note not saying it's prosumer either .

.Like I said above its pretty obvious where Focusrite was aiming this product and wishful thinking will not change that.

With that said it's probably a great sounding unit as are the Thunderbolt Ensemble and Blackface Apollo. 10 years ago people would have killed for conversion quailty at these levels for this kind of money.

Before I pack this back up I tried to look into your question...
It appears you maybe can route inputs to outputs.... Does this screenshot help?

Yes it does thanks, and does the option to route those inputs grey out once they are assigned to an output?

The next question is wether there is a matrix limit? there is one stated in the clarett documentation.

To the best of my seeing here the inputs do not grey out as you use them. Put another way you could have input 1-2 go to many outputs if you wish.

Same seems to apply to routing DAW outputs to 4PRE outputs

With the custom mix option apparently eight is the limit of these available - after eight is says the "Max Mixes has been Reached".

On another note....
I worked some more with this box and Logic today as well as PT through TB (which until all the noise blast issues are done kind of scares me even with 12.5). Got things rocking a bit better so may hang on to this one through the week after all.....

Great thanks for your continued feedback.

one last thing to try if you have a chance - Can you try routing all the ADAT and Analog inputs to Dante outputs, And visa versa - All Dante inputs to Analog/ADAT outputs. All at the same time.

In the clarett documentation for focusrite control there is a limit of how much routing you can do.

For the Clarett from focusrite control documentation
Maximum simultaneous mixing capacity:
30 inputs to
16 outputs

Just trying to clarify this for this unit.

I routed every output from some Analog, or digital input and never reached a limit. A matrix would be easier to look at - here you have to do every one individually. Regardless - it seems no limit to how much of this kind of thing you can do...

You now owe me $1 US - just PayPal it to my email... LOL...