Upgrading convertors

  • Thread starter Thread starter kenoflife
  • Start date Start date
K

kenoflife

New member
I've kind of asked this before...but wanted more of an airing of the subject...

I've been considering upgrading the convertors and clock of a Motu 828mk2
with the mod done by Black Lion Audio - its about $375 (incl. shipping) to do this - and is supposed to bring it up to the sound of RME/Metric, etc.

http://www.blacklionaudio.com/motu_firewire_mod.html

What is the relative importance of doing this? Compared to other issues like mic selection, room treatment, mic pres, etc. I know its all important, but is this more of a baseline importance - rather than the coloration of sound difference between, say a DMP3 and a Great River pre, or a Oktave MK012 vs. a MXL 603? Is this a fundamental difference that, if undone, will only allow any recording to be somehow wavery, blurred, and less than stellar? (sound of hands wringing).... :eek:
 
I would say this falls behind in priority to room/mic/pres/placement but absolutely makes a difference. Not a small one either. In this realm I have noticed more of a dramatic difference between budget gear and expensive gear. All the cheaper conversion gear I have heard sounds grainy, lacking detail, and has a smaller soundstage than the expensive stuff. That is one area where you definitely get what you pay for. Of course new converters come out every week, and upgrades are more often...
 
BigRay said:
I would say this falls behind in priority to room/mic/pres/placement but absolutely makes a difference. Not a small one either. In this realm I have noticed more of a dramatic difference between budget gear and expensive gear. All the cheaper conversion gear I have heard sounds grainy, lacking detail, and has a smaller soundstage than the expensive stuff. That is one area where you definitely get what you pay for. Of course new converters come out every week, and upgrades are more often...
Click to expand...

I agree, except that I think it makes a bigger difference than pre's, and maybe even mic's (at least you won't appreciate your mics as much until you have good converters). BTW - I had 828mkII's, and went to Apogees - best gear purchase I have ever made............
 
Ive heard one of those mods BTW, and hearing it next to a Metric Halo it still sounded like a 828mk2. There are good converters that arent THAT expensive though...Myteks are cheaper, Aurora, RME...I use myteks for my work(Classical location recording) and find them phenomenal. Save up for the good stuff.
kenoflife said:
I've kind of asked this before...but wanted more of an airing of the subject...

I've been considering upgrading the convertors and clock of a Motu 828mk2
with the mod done by Black Lion Audio - its about $375 (incl. shipping) to do this - and is supposed to bring it up to the sound of RME/Metric, etc.

http://www.blacklionaudio.com/motu_firewire_mod.html

What is the relative importance of doing this? Compared to other issues like mic selection, room treatment, mic pres, etc. I know its all important, but is this more of a baseline importance - rather than the coloration of sound difference between, say a DMP3 and a Great River pre, or a Oktave MK012 vs. a MXL 603? Is this a fundamental difference that, if undone, will only allow any recording to be somehow wavery, blurred, and less than stellar? (sound of hands wringing).... :eek:
Click to expand...
 
Converters are one area where I agree whole-heartedly, 100%, no bones about it....... save up for the good stuff. It's not like mics and pres, where it's great to have several flavors, and several mid priced items are keepers. With converters, incremental upgrades are a waste of money, and hardly noticeable. For most recordists with limited upgrade money, it probably makes the most sense to invest in a gold channel pair of A/Ds and D/As, rather than spreading that same money over 8 channels of mid grade. But of course, everybody's situation is different, YMMV.
 
Interesting....

You see, I've had (just a little more) money than time to get into all this recording stuff - and so I've been trying to build it up for when I'm really
'ready' to a certain level....
What I've gotten so far is a pair of Oktava MC12, SM-57, AT 4040, AT 4051a,
DMP-3, KRK V6, a G4 barely holding on, Logic express/Live, (just found a used) Safesound P1, silver cords, room preparations as to my recording friends specs, instruments I like, and the Motu 828MK2.....

It's about a month or two before I come out from under the work/kid/band projects a bit....so my mind goes to "is there a weak link in this chain"? Should I get the convertor/clock upgrade together so that everything is as good as it can be at this level - so I won't record that killer old Gibson guitar track, or my someday famous friends singing and be sorry later? What's another $350 at this point? Or is it already at a certain "level" that is good for awhile - waiting to be upgraded hugely if this works out (i.e. to Apogee, post-$1000 pre's, etc).....???
 
I think the converter upgrade would definitely bring out the best in what gear you have. you have the mics. Lots of albums have been recorded with less.

If your room is squared away, upgrade the converters, pre, then clock. I think youd hear a pretty big difference , hearing a "professional" converter.

Im a Mytek/Aurora guy, but lots of people like Apogee too.
 
Hey, I thought you each were each going to contribute a $1 to my Paypal account for this!
 
NL5 said:
I agree, except that I think it makes a bigger difference than pre's, and maybe even mic's (at least you won't appreciate your mics as much until you have good converters).
Click to expand...

One vote for this as being nonsense. It's like saying that the mic you use to record an electric guitar is more important than the guitar and amp you use.

Even run-of-the-mill "pro-sumer" converters these days are quite workable. Great mics and great pres into mediocre converters will always give better results than mediocre mics and mediocre pres into great converters. The greatest gains are always acheived closer to the front of the chain - song, perfomer, instrument, room, mic, pre, then converter.
 
edit - ahhhhhh, I see we have had this discussion before.....comments removed..... :D

Source is #1, and preamps are highly overated - enough said.
 
Last edited:
My opinion on converters: really, these should be the simplest things in the chain to figure out. All the cheap-to-mid-to-mid-high grade converters do is take line level signals, run them into an all-in-one converter chip (clock, PLL, filters, converter, etc.), and then send them off via various data formats, which are also driven by prefab chips. If they haven't screwed up the analog front end or power supply, the performance is driven largely by the specs of the converter chip.

They are also comparatively easy to test; since they should ideally be transparent, a D/A/D loop to see if a file nulls will do the trick.

The high end stuff, I dunno. Discuss amongst yourselves.
 
mshilarious said:
They are also comparatively easy to test; since they should ideally be transparent, a D/A/D loop to see if a file nulls will do the trick.
Click to expand...

That sounds like an interesting test. How would you set something like that up? I guess you could run your output to an input, record onto a separate track and then compare the waves. But it'd be tough to compare them precisely - what do you mean by see if the file nulls?
 
JeffLancaster said:
That sounds like an interesting test. How would you set something like that up? I guess you could run your output to an input, record onto a separate track and then compare the waves.
Click to expand...

Yes. Note that this is a combined test of D/A and A/D. Ideally you would have a reference D/A (if testing A/Ds) like a Lavry or something, but in real life the combined test is good enough.

But it'd be tough to compare them precisely - what do you mean by see if the file nulls?
Click to expand...

You record the input of your converter, line up the first samples, reverse polarity, and mix. Then you analyze the resulting mix to see how loud it is, and which frequencies are showing up as different.

Another test is a high frequency sine wave--18kHz is a fairly brutal test for a converter at 44.1; try some lower frequencies too. Plot the incoming sine wave on a high resolution FFT. Any errors introduced into the signal should ideally be far below audible levels.

Finally you can use white noise and plot on an FFT, this gives you a good idea of the filtering behavior of your converter.

Here is an example with my converters and an 18kHz sine wave. Note that I reduced the level of the test wave for clarity on the graph, and I had normalized the incoming wave to 0dBFS. This was necessary because I used a -10dBV output and a +4dBu input on the RME, because when set to the same level, the RME clipped itself :confused: Curious, but not really a problem in a recording situation.

As you can see, the original wave had a few distortions of its own, faithfully reproduced by the RME. There are smaller fluctuations in between those peaks introduced by the RME, but those are below -110dB. The filtration behavior of the RME is evident above 21kHz.

My old converters, Alesis AI3s, didn't perform anywhere near this well in the same test.
 
Hey, testing is fun! Here's my attempt at nulling a file; this is a song which I got to drop an average of -20dB or so, with the lows down -32dB and the highs not really affected much. So my conclusion is that there must be some frequency-dependent smearing caused by the trip out & back.

However, if I compare the original file with the D/A/D file, I cannot hear a difference (which of course could be a function of the D/A being used when listening), nor is the difference apparent if I simply plot the two files on the FFT.

What does it mean? In isolation, not much, but if I had two converters to compare, it might be interesting.

Again I know my old converters performed badly on this test; I could hear and plot the difference. I don't remember if I tried nulling the files though.
 
Did you line the waves up timing wise? It would have to be a couple ms off.
 
Sillyhat said:
Did you line the waves up timing wise? It would have to be a couple ms off.
Click to expand...

Yes, at the individual sample level. So the difference is more than just latency.
 
This is getting a little technical for me,
are you going to contribute $1 to my ebay account so
I can buy this upgrade, or decide I don't need it and
donate the $1 to a progressive political cause instead? :confused:
 
mshilarious said:
Yes. Note that this is a combined test of D/A and A/D. Ideally you would have a reference D/A (if testing A/Ds) like a Lavry or something, but in real life the combined test is good enough.



You record the input of your converter, line up the first samples, reverse polarity, and mix. Then you analyze the resulting mix to see how loud it is, and which frequencies are showing up as different.

Another test is a high frequency sine wave--18kHz is a fairly brutal test for a converter at 44.1; try some lower frequencies too. Plot the incoming sine wave on a high resolution FFT. Any errors introduced into the signal should ideally be far below audible levels.

Finally you can use white noise and plot on an FFT, this gives you a good idea of the filtering behavior of your converter.

Here is an example with my converters and an 18kHz sine wave. Note that I reduced the level of the test wave for clarity on the graph, and I had normalized the incoming wave to 0dBFS. This was necessary because I used a -10dBV output and a +4dBu input on the RME, because when set to the same level, the RME clipped itself :confused: Curious, but not really a problem in a recording situation.

As you can see, the original wave had a few distortions of its own, faithfully reproduced by the RME. There are smaller fluctuations in between those peaks introduced by the RME, but those are below -110dB. The filtration behavior of the RME is evident above 21kHz.

My old converters, Alesis AI3s, didn't perform anywhere near this well in the same test.
Click to expand...

Thanks, this is interesting stuff. I think I'll try some testing like this, if for nothing else than a learning experiment. Hopefully, it won't make me too dissapointed in my m-audio converters, though, as I'm not sure I'm ready to be forkin out the cash for Lavry's or something right now!
 
You must log in or register to reply here.
Back
Top