Ampex ATR 102, Anamod ATS-1, UAD, Waves - Processed Files
#61
10th October 2011
Old 10th October 2011
  #61
Lives for gear
 
Fleaman's Avatar
 
Joined: Jun 2002
Location: Los Angeles, Silverlake
Posts: 4,708

Quote:
Originally Posted by Brad McGowan View Post
I personally don't like the "zero" point chosen for the emulations, and find it difficult to adjust for a response that feels more familiar to me without resorting to an outboard EQ.

Brad
You did say your tape decks were modded, correct?
#62
10th October 2011
Old 10th October 2011
  #62
Lives for gear
 
Brad McGowan's Avatar
 
Joined: Jan 2003
Location: San Jose, CA
Posts: 6,049

That is correct. But I recall being able to get them flatter than the Anamod allows me to do before I modded them. The ATS-1 is what it is and it sounds great. I just wish I could tweak it to be flatter.

Here's some points of reference...it's definitely possible to not have the high end falling off at 10kHz with a professional tape deck:

Response Curves of Analog Recorders

I think some response plots of the ATS-1 would help illustrate what I'm talking about. I'll try to do this sometime this week. Sorry for the delay guys. It's been really busy at work.

Brad
#63
10th October 2011
Old 10th October 2011
  #63
Lives for gear
 
Fleaman's Avatar
 
Joined: Jun 2002
Location: Los Angeles, Silverlake
Posts: 4,708

Did you say there wasn't much (or any?) change in the high freq response between 15 and 30ips on the anamod?
#64
11th October 2011
Old 11th October 2011
  #64
Gear Head
 
Kwinn's Avatar
 
Joined: Aug 2008
Posts: 48

Quote:
Originally Posted by Brad McGowan View Post
I think some response plots of the ATS-1 would help illustrate what I'm talking about. I'll try to do this sometime this week. Sorry for the delay guys. It's been really busy at work.
see post #73 and #74 for some great response plots of the ATS-1:

Ampex 351, real and simulated and other tape machines

Quote:
Originally Posted by Fleaman View Post
Did you say there wasn't much (or any?) change in the high freq response between 15 and 30ips on the anamod?
30 ips and 15 ips seem to be almost identical for every machine. 7.5 ips rolls off earlier...
#65
11th October 2011
Old 11th October 2011
  #65
Lives for gear
 
Animus's Avatar
 
Joined: Feb 2005
Location: Durham, NC USA
Posts: 9,620

You should do more files but this time don't label them and see how the golden ears' expectations affect their judgments.


Trakworx
Thread Starter
#66
11th October 2011
Old 11th October 2011
  #66
Lives for gear
 
Trakworx's Avatar
 
Joined: Apr 2008
Location: San Francisco, CA

Thread Starter
Quote:
Originally Posted by Kwinn View Post
see post #73 and #74 for some great response plots of the ATS-1:

Ampex 351, real and simulated and other tape machines



30 ips and 15 ips seem to be almost identical for every machine. 7.5 ips rolls off earlier...
That is a great thread about the Anamod! Thanks for the link!

"Tape means never having to say you're sorry." - bob kats - This is going in my sig.

Brad - check out post #54 in that thread for some observations relevant to some of yours that you haven't brought up here yet.

J~
Trakworx
Thread Starter
#67
11th October 2011
Old 11th October 2011
  #67
Lives for gear
 
Trakworx's Avatar
 
Joined: Apr 2008
Location: San Francisco, CA

Thread Starter
Quote:
Originally Posted by Brad McGowan View Post
I will say this--the consistent roll-off at 20 kHz probably isn't a bad thing for recording digitally. Do you really need all that 18 kHz in your recordings? Probably not.

Oh snap! Are we just gonna let him diss 18kHz like dat?


#68
11th October 2011
Old 11th October 2011
  #68
Lives for gear
 
YOHAMI's Avatar
 
Joined: Nov 2010
Posts: 800

Quote:
Originally Posted by Trakworx View Post
Oh snap! Are we just gonna let him diss 18kHz like dat?


I thought the same. My 18khz! my 18khz!!!! dont touch my 18khz!!!!!
#69
11th October 2011
Old 11th October 2011
  #69
Lives for gear
 
TornadoTed's Avatar
 
Joined: Nov 2005
Location: Wales
Posts: 1,547

Interesting to hear other peoples findings. I just got an ATS-1 a few weeks ago and while I haven't had the chance to use it much I have ran a few things through it. My initial impression was that I didn't like the 30 IPS setting as much as the 15 IPS. I should finally get some decent time with it in the next week or so to go a little deeper and do a load more listening. I lost the little bit of paper with the card order so I didn't know which cards were in which slots. I figured I would leave finding out until after I had a good listen so I wouldn't have any pre-conceived bias!
Trakworx
Thread Starter
#70
11th October 2011
Old 11th October 2011
  #70
Lives for gear
 
Trakworx's Avatar
 
Joined: Apr 2008
Location: San Francisco, CA

Thread Starter
Quote:
Originally Posted by TornadoTed View Post
Interesting to hear other peoples findings. I just got an ATS-1 a few weeks ago and while I haven't had the chance to use it much I have ran a few things through it. My initial impression was that I didn't like the 30 IPS setting as much as the 15 IPS. I should finally get some decent time with it in the next week or so to go a little deeper and do a load more listening. I lost the little bit of paper with the card order so I didn't know which cards were in which slots. I figured I would leave finding out until after I had a good listen so I wouldn't have any pre-conceived bias!
When you're ready to find out, you can see the names on the cards and the slot numbers through the vents in the top of the ATS-1.

I prefer 15 IPS too
#71
11th October 2011
Old 11th October 2011
  #71
Lives for gear
 
toneguru's Avatar
 
Joined: Jul 2006
Location: San Francisco/LA
Posts: 2,021

Justin, you rock!

You will probably be my first call for any sessions if the need should arise. You got my kind of gear and a great attitude.

We should get together for a bite or a drink.

Stay cool and thanks for all the great work you have done.
Trakworx
Thread Starter
#72
11th October 2011
Old 11th October 2011
  #72
Lives for gear
 
Trakworx's Avatar
 
Joined: Apr 2008
Location: San Francisco, CA

Thread Starter
Thanks Tone!

I'll PM you.

J~
#73
11th October 2011
Old 11th October 2011
  #73
Lives for gear
 
YOHAMI's Avatar
 
Joined: Nov 2010
Posts: 800

Justin is the man. Brad is also the man. But Im the man.
#74
11th October 2011
Old 11th October 2011
  #74
Lives for gear
 
Brad McGowan's Avatar
 
Joined: Jan 2003
Location: San Jose, CA
Posts: 6,049

Quote:
Originally Posted by Kwinn View Post
see post #73 and #74 for some great response plots of the ATS-1:

Ampex 351, real and simulated and other tape machines


30 ips and 15 ips seem to be almost identical for every machine. 7.5 ips rolls off earlier...
Thanks for posting that link! This is pefect and will save me some trouble. This is what we noticed as well--the 15 and 30 ips settings had the same top end response...which was not what I was expecting. The A800 plots explain why I hear that emulation as boxy when all bias/EQ is set to zero. I'm going to study these in more detail. I love this crap.

I think the plots I want to now generate will focus on the effect of hitting the "tape" with different levels of signal. On my real MCI machine I notice a change in the high frequency response that is dependant upon input level--it looks like a frequency dependant low pass filter. I suspect this is what we hear as high frequency limiting/saturation.

Brad
#75
11th October 2011
Old 11th October 2011
  #75
Lives for gear
 
Brad McGowan's Avatar
 
Joined: Jan 2003
Location: San Jose, CA
Posts: 6,049

Quote:
Originally Posted by Trakworx View Post
Brad - check out post #54 in that thread for some observations relevant to some of yours that you haven't brought up here yet.

J~
Yep! This is exactly along the lines of what I just posted. I have plots from my MCI I can share that will illustrate. Here's a direct link discussing the HF limiting behavior of tape as a function of input level:

Ampex 351, real and simulated and other tape machines

It looks like those plots posted in the other thread were lacking the ATR102 card. Maybe I should run some of those.

Brad
Nrt
#76
12th October 2011
Old 12th October 2011
  #76
Nrt
Lives for gear
 
Joined: Mar 2007
Posts: 616

I'm probably minority here, but I liked ITB best. I was doing blind test, so I was quite surprised when I saw the answer. My preference was:

ITB > ATR102 >>> Anamod 102 > Burl AD/DA >>> A80 > UAD 102 > UAD 800 > ..... >>>>>>>>>>>>>>>> Waves (horrible)

This test clearly told me that the software emulations sound still somewhat digital, and Burl is not my taste. If the converter was Prism or Mytek, ATR and Anamod (and even UAD) would have been more preferable than ITB, I suspect. I think the combination of UAD and Burl is quite bad, since both sound somewhat smeary (warm?, maybe...). Real ATR is great as usual.

Thank you for the test!
#77
12th October 2011
Old 12th October 2011
  #77
Lives for gear
 
Brad McGowan's Avatar
 
Joined: Jan 2003
Location: San Jose, CA
Posts: 6,049

Here's the response of my MCI JH-110A 1" 8-track. There are two different calibrations shown both with the test signal reading -6 VU on the meters. The calibration settings were for GP9, 15 ips, +0/355. The overbias and HF EQ appear to have been tweaked differently for the blue trace. Also you can see I dialed in a different amount of LF EQ as well during that alignment. But the orange trace is relatively flat and extended. It's +/- 0.5 dB from 120 Hz to >20 kHz. the largest bump in the low end is only 1.5 dB. Between the two traces you can see how one can flatten out or accentuate the low end. If I live with a little dip at 65 Hz then I can probably tame the 40 Hz bump a little more.

My point is that tape decks can be very flat and extended in the high end at 15 ips. Mine certainly is.

Brad
Attached Thumbnails
Ampex ATR 102, Anamod ATS-1, UAD, Waves - Processed Files-mci-6vu-baseline.jpg  
Trakworx
Thread Starter
#78
12th October 2011
Old 12th October 2011
  #78
Lives for gear
 
Trakworx's Avatar
 
Joined: Apr 2008
Location: San Francisco, CA

Thread Starter
Quote:
Originally Posted by Brad McGowan View Post
My point is that tape decks can be very flat and extended in the high end at 15 ips. Mine certainly is.
Too much 18k
Trakworx
Thread Starter
#79
12th October 2011
Old 12th October 2011
  #79
Lives for gear
 
Trakworx's Avatar
 
Joined: Apr 2008
Location: San Francisco, CA

Thread Starter
But seriously,

Brad, do you have a plot that shows the level dependant low pass filter effect? i.e. one with a hotter input level?

J~
#80
12th October 2011
Old 12th October 2011
  #80
Lives for gear
 
Brad McGowan's Avatar
 
Joined: Jan 2003
Location: San Jose, CA
Posts: 6,049

Here's a plot showing that as input level increases (0VU, +6VU, +10VU) the high frequency response is attenuated. I have normalized all three traces so that you can clearly see the high frequency limiting / saturation that occurs as a function of input level. This is one of the dynamic features of analog tape that gives it its characteristic smoothness and fatness. Imagine what happens to a high frequency transient on a snare drum or electric guitar signal.

I intend to do the same measurements on the Anamod to illustrate whether or not the ATS-1 emulates this behavior. Justin--I'll send you the test signal so that you can maybe run it through the plugins at various settings.

Brad
Attached Thumbnails
Ampex ATR 102, Anamod ATS-1, UAD, Waves - Processed Files-mci-hf-limiting.jpg  
Trakworx
Thread Starter
#81
12th October 2011
Old 12th October 2011
  #81
Lives for gear
 
Trakworx's Avatar
 
Joined: Apr 2008
Location: San Francisco, CA

Thread Starter
Nice. Thanks!
#82
12th October 2011
Old 12th October 2011
  #82
Lives for gear
 
Joined: Mar 2008
Posts: 803

Quote:
Originally Posted by Brad McGowan View Post
Here's a plot showing that as input level increases (0VU, +6VU, +10VU) the high frequency response is attenuated. I have normalized all three traces so that you can clearly see the high frequency limiting / saturation that occurs as a function of input level. This is one of the dynamic features of analog tape that gives it its characteristic smoothness and fatness. Imagine what happens to a high frequency transient on a snare drum or electric guitar signal.

I intend to do the same measurements on the Anamod to illustrate whether or not the ATS-1 emulates this behavior. Justin--I'll send you the test signal so that you can maybe run it through the plugins at various settings.

Brad
Hey Brad,

A word of caution here. Impulse response measurements are only valid for linear systems. You'll need to use a steady state signal and account for the distortion to make the type of measurements you're trying to make. This is most likely not the actual frequency response as is it's not level dependent. The software you're using is intended for linear (approximately linear) acoustic measurements. You may wish to use pseudo-random noise or a sweep sinewave which would give you accurate results.

Best,

/Dave
#83
12th October 2011
Old 12th October 2011
  #83
Lives for gear
 
Brad McGowan's Avatar
 
Joined: Jan 2003
Location: San Jose, CA
Posts: 6,049

Hey Dave!

Thanks for the tips. I did indeed use a log sinewave sweep for these measurements (starting at 10 Hz). The software did give me the option to use a sweep or noise signal and I chose that instead of the impulse. Ignore the impulse response plot at the bottom of the screen. I couldn't figure out how to hit that part of the display.

thanks,
Brad
#84
13th October 2011
Old 13th October 2011
  #84
Lives for gear
 
Brad McGowan's Avatar
 
Joined: Jan 2003
Location: San Jose, CA
Posts: 6,049

I just ran some measurements on the the Anamod and have been gettting response plots identical to those posted by Steffen in the other thread. So at least I know I'm getting consistent results.

What I'm not seeing is any indication that the Anamod emulations capture the dynamic high frequency limiting behavior that real tape decks would exhibit. The response plots are identical for any input level I feed it. Even If I crank the input level all the way up I still don't see any difference in response. Interesting.

What I am measuring is different distortion spectra for each tape formulation. As expected GP9 is the cleanest, then 456, and then 111, which has the highest level of distortion.

I'm also seeing the behavior of the bias control. As the bias knob is turned counter clockwise, LF distortion increases slightly, HF distortion decreases, and the frequency response flattens out in the high end. A flatter top end definitely results when cranking that bias knob to the left. As the bias knob is turned clockwise, LF distortion decreases, HF distortion increases, and the frequency response rolls off in the high end.

I'll post some plots later.

Brad
#85
13th October 2011
Old 13th October 2011
  #85
Lives for gear
 
YOHAMI's Avatar
 
Joined: Nov 2010
Posts: 800

Any chance we can get an additional round? maybe the tape / anamod on the individual instrument tracks?
Trakworx
Thread Starter
#86
13th October 2011
Old 13th October 2011
  #86
Lives for gear
 
Trakworx's Avatar
 
Joined: Apr 2008
Location: San Francisco, CA

Thread Starter
Interesting. Thanks Brad!

I'll get you some test files on the plug-ins (and HW) as soon as time permits. I'm curious to see if any of them modeled the dynamic HF response.

As an aside - I got my first-ever reel of 1/2" SM911 yesterday. Interestingly, with both 911 and ATR tape I get a flatter high end than with 456 or 499. -2dB instead of -3dB at 20k, -0.9dB instead of -1.2dB at 16k. Flatter through the mids too.

I tried 2 reels of "one-pass" 499 from Tape Tape. Both were badly worn out - trying to align at 10k was like chasing a pea around a dinner plate with a dull fork. One pass my a$$!

So glad they're making quality tape these days!

J~
Trakworx
Thread Starter
#87
13th October 2011
Old 13th October 2011
  #87
Lives for gear
 
Trakworx's Avatar
 
Joined: Apr 2008
Location: San Francisco, CA

Thread Starter
Quote:
Originally Posted by YOHAMI View Post
Any chance we can get an additional round? maybe the tape / anamod on the individual instrument tracks?
Yes, more will be coming. Studio time is scarce lately, but soon...

J~
#88
13th October 2011
Old 13th October 2011
  #88
Lives for gear
 
YOHAMI's Avatar
 
Joined: Nov 2010
Posts: 800

Awesome
#89
13th October 2011
Old 13th October 2011
  #89
Lives for gear
 
Joined: Mar 2008
Posts: 803

Hey Brad,

What tape formulation and operating level are you using for your measurements and does 0 dB mean you are at said operating level? Are the graphs we're seeing sweep sine or pseudo random noise? Also, do you have a way of time aligning the repro signal?

Best,

Dave
#90
13th October 2011
Old 13th October 2011
  #90
Lives for gear
 
Brad McGowan's Avatar
 
Joined: Jan 2003
Location: San Jose, CA
Posts: 6,049

Dave,

Are you referring to the plots I posted of my MCI machine? Those plots were made using GP9 at 15 ips. The test signal was a sine wave sweep starting at 10 Hz. The operating level for the test (which was done a year ago) was relative to 0VU = -15 dBFS. The deck at that time was aligned to 0/355. I varied the test signal output from the computer by changing the level (within the HOLMImpulse software) relative to -15 dBFS. Is that what you were asking?

As for the tests I'm running on the ATS-1... I'm not exactly sure what the converters are calibrated to at home but I'm in the ballpark of -16 dBFS for my test signal. I say ballpack because the audio interface has some input and output trims. I'm calling that my 0VU point. I left the input knob at 12:00. So I did not adjust the input to see 0VU on the ATS-1's meters.For the quick tests I ran this morning I tried a couple different machine cards and GP9 I believe. I can document and calibrate things more carefully a little later tonight using a multimeter. Again, I'm using sine wave sweeps.

Let me ask you this: if I feed a test signal at 0VU = 1.23VAC into the ATS-1, and I set the meter to +6, should I turn up my test signal by +6 dB to maintain the appropriate virtual operating level / flux that would correlate to what one would get with a +6 alignment? Does my question make sense? What amount of gain boost/cut does the input knob give you? on the ATS-1?

I'm not sure I understand what you mean when you ask if I have a way of time aligning the repro signal. Time align it relative to what? Care to elaborate on what you are thinking?

Any help you can offer in testing accurately is greatly appreciated. I welcome your expertise! Let me know if what I described above regarding the distortion of the tape types and the bias control is correct. If I'm getting erroneous results due to my testing methodology then I'd like to correct it.

Justin--looking forward to getting those files back from you on the plugins.

Thanks!
Brad
New Reply Submit Thread to Facebook Facebook  Submit Thread to Twitter Twitter  Submit Thread to LinkedIn LinkedIn  Submit Thread to Google+ Google+ 
 
Thread Tools
Search this Thread
Search this Thread:

Advanced Search
Forum Jump
 
Register FAQ Search Today's Posts Mark Forums Read

SEO by vBSEO ©2011, Crawlability, Inc.