Orpheus - pulling the plug for location work?

[neirbod]

Hello all,

I have been using my trusty Fireface 800 for many years for location recording. I think it sometimes gets a bad rap around here, but I find that as long as I use external pres and record at 96 khz it actually stands up quite well. That being said, I got an Orpheus last week to demo. Using the same external pres to compare apples to apples, I find the Orpheus converters to be subtly but singificantly superior - tighter lows (most noticeable in kick drums), more transparent, and with a 3D-like quality I never heard from the FF. I am sold on the benefits of higher quality conversion, but not sold at all on the Orpheus as I question how reliable it is for location work.

Let me elaborate. As fellow location recorders, I don't need to sell anyone here on the need for a reliable backup, especially if your main system relies on a laptop and firewire connection. With my FF, I record multitrack over firewire to a laptop as "plan A." For "plan B" I make a stereo mix within the FF mixer and route this to a portable stereo recorder via the S/PDIF output. Works like a charm and has saved my butt a couple of times. Importantly, if I listen to what is coming into my backup recorder and pull the firewire plug to simulate a signal loss in my main system, the backup mix continues perfectly with no audible glitch.

Now we come to the Orpheus. I tried the same backup system and simulated a signal loss. The Orpheus did continue to pass the audio to the backup, but only after 2-3 seconds of a complete dropout. I tried this listening to the S/PDIF out, the analogue outs, and the ADAT outs (in direct mode) and had the same dropout each time.

Has anyone else tried this? I am curious if my experience is unique, or if this is just the way the Orpheus works. If the latter, this could be a deal breaker as I can't think of a reasonable workaround. Standalone mode would presumably work, but then you can't adjust gain settings.

I love the sonics of the Orphues, and the lightweight 1U solution that combines nice pres, great converters, and overkillers. So, I would love to be shown the error of my ways or an elegant workaround so I could use it for location work. But, I am drawing a blank.

Thoughts?

[John Willett]

Originally Posted by neirbod

Thoughts?

Copy this to Prism direct and ask the question.

It's an important point that they need to be aware of.

People often comment on public forums and don't talk to the manufacturers - and if the manufacturers don't know, they can't fix or change things.

[John Willett]

Originally Posted by John Willett

Copy this to Prism direct and ask the question.

In fact I have now done this for you.

I have copied this to Prism so they are aware of it.

This really *is* the sort of thing that a manufacturer needs to be aware of.

[rumleymusic]

Has anyone else tried this? I am curious if my experience is unique, or if this is just the way the Orpheus works.

I have been concerned about this issue with a number of FW interfaces. I have owned a Mackie 1200F, Motu 828mk3, and Steinberg MR816x. So far, the only one that remains rock stable after a complete computer restart is the Mackie. Both the MOTU and Steinberg will cycle through a startup sequence and cut the audio to the outputs for a couple of seconds. They both, however, will remain stable if the FW plug is disconnected.

I think a company as high end as Prism should take this issue seriously, even if it means a complete driver overhaul.

[neirbod]

John,

Thanks for stepping in. For the record, my approach was to first see if others could replicate the problem to see if this was an isolated issue with my system vs. a systemic problem. Obviously, depending on how widespread this issue is will inform any future conversation I, or others, may have with them. I was then going to contact Orpheus having a bit more information to go on. But, no harm done in you contacting them now. Please pass on what you learn.

Also, I had spoken with Frank Oglethorpe at Prism prior to getting the unit and asked about the Orpheus' behaviour in the event of a firewire signal loss. He indicate he thought it would still pass audio and digital signals without issue, but suggested I try it myself. I found it a little strange they didn't have a more concrete answer to my question as this is a very significant issue I thought they would have explicitly tested.

I am still curious if others are able to replicate this problem. It is very easy to check, so I hope people do it and report back. I hope I am the exception to the rule, and I just need to tweak something.

Dave

[neirbod]

Originally Posted by rumleymusic

ISo far, the only one that remains rock stable after a complete computer restart is the Mackie.

Add Fireface to the list of cards that preform admirably.

Originally Posted by rumleymusic

I think a company as high end as Prism should take this issue seriously, even if it means a complete driver overhaul.

Agreed 100%.

I need to point out that I really don't want to start spreading false info. As far as I know at this point, it is an issue isolated to my system. I am very curious what others may find.

[John Willett]

Originally Posted by neirbod

John,

Thanks for stepping in. For the record, my approach was to first see if others could replicate the problem to see if this was an isolated issue with my system vs. a systemic problem. Obviously, depending on how widespread this issue is will inform any future conversation I, or others, may have with them. I was then going to contact Orpheus having a bit more information to go on. But, no harm done in you contacting them now. Please pass on what you learn.

Dave

I forwarded the text and a link to this thread to a technical guy at Prism who will forward it to whoever needs to know within the company.

I expect they will reply direct on this thread.

If they reply to me, I will post what they say here.


Posted via the Gearslutz iPhone app

[neirbod]

I have had a couple of email exchanges with Frank at Prism about this. I am impressed by their customer support - Frank has been very responsive to my emails and phone calls.

Frank suggested I try to simulate a system crash by turning off the computer instead of pulling the firewire cable as the result may be different. So, I cold rebooted the computer by holding down the power button and found essentially the same result. The Orpheus audio dropped out. The dropout was shorter, about a second, compared to 2-3 seconds when I pulled the firewire cable. Still, a second of dropout is a second too long.

So, anyone else try this? With all the Sultz out there raving about the Orpheus, I thought this discussion would have raised more interest. I am *very* curious to know if others can replicate these dropouts.

Dave

[John Willett]

Originally Posted by neirbod

I have had a couple of email exchanges with Frank at Prism about this. I am impressed by their customer support - Frank has been very responsive to my emails and phone calls.

Frank suggested I try to simulate a system crash by turning off the computer instead of pulling the firewire cable as the result may be different. So, I cold rebooted the computer by holding down the power button and found essentially the same result. The Orpheus audio dropped out. The dropout was shorter, about a second, compared to 2-3 seconds when I pulled the firewire cable. Still, a second of dropout is a second too long.

So, anyone else try this? With all the Sultz out there raving about the Orpheus, I thought this discussion would have raised more interest. I am *very* curious to know if others can replicate these dropouts.

Dave

I sent my e-mail to Simon Woollard at Prism - he forwarded it to whoever needed to know - so they are definitely taking this seriously.

Their Customer support is excellent.

[neirbod]

An email I got from Prism today confirms that my situation is not unique:

"Our R&D team concurred that you would get a brief drop out on [computer]
shut down which I'm told is down to the Orpheus firmware and that it re-starts streaming so that it can synchronize to its firewire bus. So it's unlikely we'll be able to alter this in the near future..."

This is unfortunate, obviously. For me, this is a deal-killer. I hope Prism takes a vary hard look at their firmware and brings it up to par with the Orpheus hardware.

On to other options, the best bet at this point appearing to be a Mytek 8X192 feeding my good ol' Fireface over ADAT.

[panatrope]

I have trialled an Orpheus and the performance and a lot of its features are very nice (like the SRC - record 24/96K multitrack, record a safety CD from a dithered 16/44K1 SPDIF output). There are some other things that Prism could do to the Orpheus to make it more production friendly for broadcast, one of which is to provide access to the ADAT in and out in the mixer. These channels cannot presently be used as part of an ITB mix.

However, looking at the system as a whole, you do also have to ask about the level of risk of processor outage during recording. And that is something that the user also has some control over. 'Hardening' the PC host may be more readily achieved than waiting for Prism to produce a reliable firmware upgrade.

If an on-the-spot 'live' or safety mix is required, then maybe a different configuration is better. Think about a preamp like the Micstacy, where the recording signal goes via ADAT or AES links, while the on-the-spot mix is conducted from the line outs using your mixer of choice (an X-Desk could be fun) to a stand-alone recorder (or your broadcast feed). No matter what happens on the computer side, the Micstacy doesn't stop amplifying signals ....

Maybe Prism could pick the brains of their new SaDIE chums when contemplating Orpheus V2.0

[David Spearritt]

Originally Posted by neirbod

This is unfortunate, obviously. For me, this is a deal-killer. I hope Prism takes a vary hard look at their firmware and brings it up to par with the Orpheus hardware.

And for me as well. I loved the "look" and sound of the Orpheus from the demo unit I had for a few days, but paying $5K for a FW device having limited life, as well as possible glitches when stressed under typical location conditions, it all causes hesitation and ambivalence. I think FW will die soon, and its constant tweaking and experimentation with cheap crappy cards and cables is a complete joke for classical recording. Not to mention the hopeless physically dodgy express card constraints in the laptop itself.
RME User Forum / Problems mounting ExpressCard into a Laptop

I like Panatrope's thinking, and this is what I still use for critical gigs. Dual system, analog mixer (wonderful X-Desk) feeding stereo master recorder, and X-Desk raw outs going to digital multitrack.

If only Nagra would make a 12 channel recorder ... Sigh.

[tuRnitUpsuM]

Out of curiosity why not just run a completely isolated back up ? You are asking 1 firewire device to stream 2 separate groups of signals... 1 (multi-track) and 1 (stereo). Thats alot to ask the drivers to keep track of. Especially in Real-Time during a brief moment of high-stress (an abrupt shutdown or power loss). Surely the "improvements" you hear with the Orpheus would warrant a change in workflow? again just curious....

cheers

[panatrope]

Originally Posted by tuRnitUpsuM

Out of curiosity why not just run a completely isolated back up ? You are asking 1 firewire device to stream 2 separate groups of signals... 1 (multi-track) and 1 (stereo). Thats alot to ask the drivers to keep track of.

Not really at the track counts (6 - 10) I am talking about, and even around double that number, we are well clear of the FW transmission limit. The drivers are not usually the issue

But the stereo feed we are talking about here is an independent hardware one (on AES or SPDIF) straight out of the box, not going to the host. So FW/driver capacity does not come into it. (And of course, using the FW to carry the multitrack and stereo feed would defeat the purpose of the exercise because if the host goes down, the multitrack AND the stereo feed is lost anyway).

The Orpheus is perfectly suited to studio-type rehearse record sessions - if the host has a hiccup, just start a new take when it is recovered. For live recording or as part of a broadcast chain, then you have to take the risk of host failure into account.

BTW, another 'remote' disadvantage - unlike most of my core gear, the Orpheus will only run on mains with no provision for battery operation. So if some other person does something inappropriate and trips the breaker that is feeding my gear, everything but the Orpheus will survive. Nagra, Traveler/RME, laptop, standby bitbucket - all can be run on battery alone.

[sonare]

I have been using Orpheus for several years as 8/16/24 channel devices and have not had a problem even remotely similar to what you describe. I feed the HD24 backup from the analog outpit of the Orph and the SPDIF/AES supplies the monitoring.

There are certain realities of using a computer for live work that causes one to either minimize the potential failure modes or choose a different way of working. With my signal chain the Orphs maintain analog signal output even if the computer bites the dust. How could I ask for more? Hardening against an EMP weapon?

Rich

PS You DO know about UPS devices, don't you?

[marvin100]

Originally Posted by John Willett

In fact I have now done this for you.

I have copied this to Prism so they are aware of it.

This really *is* the sort of thing that a manufacturer needs to be aware of.

Agreed entirely.

But I also think responsive manufacturers should be running RSS feeds and notifications on all relevant boards so they know when their products are being discussed. It's so easy and takes so little time/effort/expertise, I'm shocked if it isn't standard procedure at this point.

[tuRnitUpsuM]

Originally Posted by panatrope

Not really at the track counts (6 - 10) I am talking about, and even around double that number, we are well clear of the FW transmission limit. The drivers are not usually the issue

But the stereo feed we are talking about here is an independent hardware one (on AES or SPDIF) straight out of the box, not going to the host. So FW/driver capacity does not come into it. (And of course, using the FW to carry the multitrack and stereo feed would defeat the purpose of the exercise because if the host goes down, the multitrack AND the stereo feed is lost anyway).

The Orpheus is perfectly suited to studio-type rehearse record sessions - if the host has a hiccup, just start a new take when it is recovered. For live recording or as part of a broadcast chain, then you have to take the risk of host failure into account.

BTW, another 'remote' disadvantage - unlike most of my core gear, the Orpheus will only run on mains with no provision for battery operation. So if some other person does something inappropriate and trips the breaker that is feeding my gear, everything but the Orpheus will survive. Nagra, Traveler/RME, laptop, standby bitbucket - all can be run on battery alone.

There is no such thing as an independant hardware output on an all-in-one device. The hardware outputs (in this case the AES-SPDIF digital outs) are still controlled by the units software mixer (which in turn is still controlled by the units driver). A driver does not necessarily have to talk to a host (usually a computer) to be active.....its always active so long as the unit is powered. Without it... you just have physical connectors and software connections (mixer). It is the brain of the unit - the driving force. It is however mostly used in the context of "talking" to the hosts (ie computer protocol - example Firewire) which is a misconception of what it actually is responsible for . At least that is my understanding of how interfaces work. Of course i could be wrong here and the Orpheus actually has separate "drivers" to actually give you an independant output.

Think of it like this..... the Orpheus becomes the pistons ,rings, gasket, crankshaft etc in a combustion engine.....while the drivers would be the equivalent of the gasoline and air mixture, the result of the compression and burning of that mixture, the oil etc. Without either , neither function.

Back to the question imposed..... if you ( OP - anyone else in this situation) not necessarily just you.... run an entirely separate back-up. An example a Sound devices 702 or similar (anything to record an audio signal in stereo)... you avoid all these complications. You have a TRUE redundant back up.

cheers

[neirbod]

Panatrope and Sonare - if I understood your posts, you both have backup systems that rely on Orpheus continuing to pass a stereo signal in the event of a system failure. This was my plan, until I discovered the problems. I urge you to check what happens in the few seconds immediately following a simulated failure, unless a temporary dropout does not pose a major problem for you.

Let me reiterate that the FF does reliably pass audio glitch free. For me, keeping the FF as a hub and bringing in external coverters (I have my eye on the Mytek 8x192) seems to make the most sense, although I am still mulling options.

For those suggesting use some kind of split prior to hitting the orpheus, that is of course one approach. But, that precludes using the Orpheus' pres, which limits its benefits.

I hope Prism figures this out with a new firmware, but I certainly am not waiting for that. While they are at it, a more flexible and intuitive mixer, including being able to mix in ADAT channels, would be very helpful.

[sonare]

Re-read my post-- I do not rely on Orph for a digital stereo anything involving backup. Assuming the Orphs and HD24 continue to get AC from the UPS the analog output from Orph to HD24 continues regardless of what the host computer does. I have tested this. I cannot imagine any device that would continue to send ADAT if communication between interface and driver is lost-- and I decided early on NOT to use the ADAT outputs due to my need to monitor thru my DAW (Sequoia)-- the channel numbering in the ASIO driver changes when dong that. Everything is more reliable going analog into the HD24 rather than ADAT. And I am not trying to accommodate SMUX for 2fs.

If a RADAR rig is not in the budget decide what you can't live without and make equipment purchases and plans to make your plan work. What you propose is a 100% reliable host computer and no such thing exists. I am satisfied with 99% with a 100% solid backup. Yes-- this assumes electrons will still flow into the gear. But even if 100% dependable power is a given how do you make sure an airplane doesn't crash into the hall?

In 8 years of relying on a host computer I have had to use a backup exactly 4 times. One of those was a session and the blitzed take was so fabulous the client wanted the backup (even though there was a difference in sound). He chose music and I agreed with him.

The Mytek sound is great-- and I hope you never have to send it in for service as I had to do. I prefer my sound of the Prisms anyway. As you make your decisions be sure you don't confuse the interface "problems" with the host "problems." Good hunting!

Rich

[neirbod]

Rich,

I think you need to reread *my* post. I never said the issue was related only to digital. I get the same temporary drop out with the analog outs. I am very curious, and your post was not specific - does your Orpheus pass analog audio *with no dropout* in the event of disconnecting from the host? The fact it passes audio is not in question, but whether it does so without a couple seconds of dropout is very much in question.

Also, the FF at least passes S/PDIF perfectly if the connection to the host is lost. Can't say for sure about ADAT. I am quite comfortable with the FF doing its job as long as I have power (and yes, I use a UPS).

Dave

[panatrope]

Originally Posted by tuRnitUpsuM

Think of it like this..... the Orpheus becomes the pistons ,rings, gasket, crankshaft etc in a combustion engine.....while the drivers would be the equivalent of the gasoline and air mixture, the result of the compression and burning of that mixture, the oil etc. Without either , neither function

Unfortunately, your analogy is more confusing than accurate.

In most interfaces, inclusing the RME and the Orpheus, the on-board processor (DSP chip, FPGA, whatever ...) serves many functions, only one of which is to manage the link from device to host (FW/USB/MADI). It also provides control and supervision of the whole device and acts as a mixer between the local inputs and outputs, controlled by the HUI interface program running on the host which sends control messages over the link. If the link is not working, it will either manage the attempts to resynchronise, or adjust the local working conditions (like turning on the red light that shows a link error). The local mixer keeps working using the last set of parameters sent over the link before it failed. Further local adjustments can usually still be made using a knob on the front panel.

The Orpheus appears to be able to do this but only after the on-board processer devotes all its time (for a few seconds) to trying to restore the link, during which time the mixer stops working. Thus ALL signals fail including the analog in to analog out because these have to pass through the mixer (there is no direct analog in-analog out path in the interface).

By comparison, devices like the Fireface appear to have arranged their on-board processor so that it continues to provide the mixer function while the link handling portion of the program works independently to restore the link. This means that outputs continue without interruption - analog, SPDIF, ADAT. The only thing that is lost is supervision and control of the mixer from the host (eg. TotalMix). This returns without interruption when the host is rebooted, and is independent of the DAW application being use.

Based on previous experience, I am currently sufficiently confident to use an interface like the Fireface as part of a live broadcast signal chain because of this operational independance. While the Orpheus exhibits the present behaviour (and lacks a couple of other operational facilities), I am more hesitant about using it in similar circumstances. If it is basically a matter of firmware, this can possibly change in the future. If however, the Orpheus uses a dedicated chip to manage the FW link, it may not be possible to change this behaviour without modifying hardware.

Note, as I previously remarked, this is only critical in certain Orpheus applications. In others, where continuity of operation is not mission-critical, I would be very happy to use the Orpheus for its audio and other qualities.

[tuRnitUpsuM]

There is no such thing as an independant hardware output on an all-in-one device. The hardware outputs (in this case the AES-SPDIF digital outs) are still controlled by the units software mixer (which in turn is still controlled by the units driver). A driver does not necessarily have to talk to a host (usually a computer) to be active.....its always active so long as the unit is powered. Without it... you just have physical connectors and software connections (mixer). It is the brain of the unit - the driving force. It is however mostly used in the context of "talking" to the hosts (ie computer protocol - example Firewire) which is a misconception of what it actually is responsible for . At least that is my understanding of how interfaces work. Of course i could be wrong here and the Orpheus actually has separate "drivers" to actually give you an independant output.

My friend, this is what you shouldve quoted... and not worried so much about the analogy. Because worrying about the analogy , you've echoed this quote.

The RME is capable of the task because it does have...separate drivers. One for Totalmix and another for the other tasks an interface must perform. Communication between host and device etc etc. It is the reason there is a delay on the Orpheus. There is one driver running the show. The RME has a hardware buffer (DSP) between it and the software. RME is known for developing the best drivers in the biz with good reason because there are two of em .

HUI ? Heads up Igor? teasing..... the analog ins and outs are also controlled by the driver...subsequently the mixer (which is controlled by the DRIVER). The mixer is only a GUI portion of the routing infrastructure... it really is just a means of viewing what is happening... everything is under direct management of the DRIVER.

Still strongly urge the use of a completely and entirely separate back up. Its the only way one can be sure of a fail safe (r) .

[panatrope]

I fear this discussion is becoming like the great Yogi's dictum "When you come to a fork in the road, you should take it". or in a more literary vein, the great George Bernard Shaw's observation about the USA and Britain: "Two great nations separated only by a common language". Basically you and I appear to have a different concept of the meaning of 'driver'.

Under my understanding, and perhaps you can check this with RME and Prism, the driver program is what resides in the host (and includes, in the case of RME, the TotalMix control surface, and in the case of Orphues, its particular mixer ... they are part of the driver). The firmware that resides in the device is the operational program for the DSP/mixer controller, which occasionally receives operating commands from the host based driver via the link (FW/USB/MADI/whatever) whenever a change in the mixer is requested via input from the user via the host based control screen.

So, in my understanding, it is not so much a driver (in the host) issue, as the way the firmware in the device responds to a link loss. RME devices are appartently programmed to handle this without interrupting the operation of the firmware mixer; the Orpheus is not.

Quite frankly, if it weren't for their ridiculous 60mm faders, a Mackie Onyx FW mixer could be an ideal choice for this in-line broadcast plus multitrack recording role. Prism designed the Orpheus the way it is, because the reliability requirements in the way I would like to use it wasn't part of the user requirement brief to the designers. If by the sort of discussion in this thread they are now influenced to think it is a valid requirement, they will modify the firmware accordingly. If they don't, it will never replace the RME in my kit, no matter how good its audio performance may be. Simple as that!

[neirbod]

Originally Posted by panatrope

RME devices are appartently programmed to handle this without interrupting the operation of the firmware mixer; the Orpheus is not.

... Prism designed the Orpheus the way it is, because the reliability requirements in the way I would like to use it wasn't part of the user requirement brief to the designers. If by the sort of discussion in this thread they are now influenced to think it is a valid requirement, they will modify the firmware accordingly. If they don't, it will never replace the RME in my kit, no matter how good its audio performance may be. Simple as that!

Agreed. Personally, I don't care much if the issue is a driver or firmware. I just would like it to work the way I need it so I can take advantage of the great sonics.

This is a very interesting and informative discussion. I hope Prism is listening.

[tmcconnell]

I used my Orpheus in a live situation and will never do so again. It could have been the FW, it could have been the computer, it could have been the power supply, but.... the system failed miserably. I have not had similar problems in the studio (well, sort of, but easy to fix with a boot). Its a sort of thing that, when its right its right, but it is also the canary in the mineshaft. If the computer sneezes, the Orpheus catches a cold. In "direct" mode, there's a lot less to fail, but after being humiliated in front of a client, I will now only use bomb proof gear with backup for everything. Cranesong spider adat out to an hd24. Sorry to say I would not recommend Orpheus with a FW laptop --- even purpose built audio laptop---- for commercial live work. In fairness to prism and the OP, the issues I had were probably not the fault of the orpheus, but when I had to repeatedly boot all components in the right order to get back into business, I gave up the idea of a primarily software driven box controlling so many basic functions. In live work, your job is to capture the bits with perfect precision. You can pick up the pieces later with your daw in a studio.

[panatrope]

Originally Posted by panatrope

If by the sort of discussion in this thread they are now influenced to think it is a valid requirement, they will modify the firmware accordingly. If they don't, it will never replace the RME in my kit, no matter how good its audio performance may be. Simple as that!

And while they are at it, support in the mixer for a control surface - in my case an MCUPro - would also put it on a par with the RME, operations-wise. Secondary perhaps to the reliability issue, but now an almost mandatory requirement in the live operations environment - in my opinion anyway.

[neirbod]

I am glad to see this thead is still going, and over 1,000 people have viewed it. Perhpas this will inspire Prism to make the necessary changes that many here have described.

I intend to return my Orpheus due to its limitations, but will first use it to compare with another well respected converter I picked up as part of my research to upgrade my FF. I am doing a choral recording in the next few days. It is in a church with decent acoustics and with no audience. I will use Grace m801 pres (with dual outs) to feed both converters from the same mics. I will try to use two separate set ups if time permits - Schoeps MK4s in ORTF; and an omni AB pair of Josephson c617s. I will keep the identity of the other converter secret for now. I'll post files in a separate thread for your listening pleasure next week.

I really do like the Orpheus sound, so I hope I don't hear a difference, or at least a preference. I will use Foobar to be as unbiased as possible, and will encourage others to do the same.

Stay tuned!

Dave

[chemelneme]

Hello Dave!

Your posts have been great! I'm actually thinking about upgrading my interface and my top choices are the RME UFX and the Orpheus.

Most people say the Orpheus sounds better and thats a big plus for me. However since you have access to both interfaces, maybe you can help me out. Which interface has better performance? For example: with which interface can you work with lower latencies and higher number of tracks/plugins without glitches? Or do they work the same? Did you have problems setting up the interfaces with your system?


I know some of you guys will shoot back: "that depends on your computer hardware". But in my experience that is not entirely true. It is also very important how the interface/driver work together. I have a MOTU Ultralite and it's been kind of tricky to get it to work properly. First I had to change the computer because the Motu didnt like FW chipset the original computer had. Also, I can only use the driver that was shipped with the interface. If I try to upgrade it to a newer version, the interface makes cracking noises (while recording or playback), because of CPU usage spikes. The drivers for PC aint that good on my interface.

Thanks for all the help.

[Sheikyearbouti]

Originally Posted by chemelneme

Hello Dave!

Your posts have been great! I'm actually thinking about upgrading my interface and my top choices are the RME UFX and the Orpheus.

Most people say the Orpheus sounds better and thats a big plus for me.

Are you sure this was said about the UFX? It is a very new interface and I believe that those, otherwise fair opinions, were expressed having in mind something like the FF400/FF800.
The UFX is supposed to have a better performance. I don't know how much better and how would it compare to the Prism but at least that is what RME states (new converters, preamps...).

[neirbod]

Originally Posted by chemelneme

Most people say the Orpheus sounds better and thats a big plus for me.

Not sure if you saw this, but it pretty much confirms that the Orphues converters were preferred over the FF800. However, nearly a third still prefered the FF800 on some clips and some who heard a "clear difference" chose the FF one one clip and the Orpheus on the other, so it was not a slam dunk by any means:

FF800 vs. Orpheus converter test

My personal opinion is that the difference is subtle, but may be significant depending on what your audience and goals are. I have used the FF800 for years and my clients love my work, but I still want to upgrade converters in the near future. I am 99% sure one of the FF units will continue be the hub of my rig, even if I need to bring in external converters, as Total Mix's flexibility and the driver stability are critical for how I work. No idea yet how the UFX will compare sound wise, but if it is close to the Orpheus I will buy one.

As for other aspects of performance, everything I know is posted above. I returned the Orpheus shortly after I did these tests since the driver stability was not good enough for how I work. Also, the internal mixer is very rudimentary compared to RME's Total Mix, which may or may not be a factor for you. For what it is worth, I can easily get down to a few ms latency on my FF800 with 12 or so tracks and multiple effects. Never had the need to push it more than that.