The No.1 Website for Pro Audio
API 1608 Mix Buss - Op Amp Blind Listening Test - API 2520, 990C, Rogue 5, GAR 2520, VF600, Red Dot
Old 28th February 2021
  #241
Lives for gear
 
🎧 10 years
If anyone has any interest here’s my first mix with my 1608 full of red dots. Straight off 16 track tape machine. 1608 2 buss into pro tools.. all analog processing no editing or punches

Old 1st March 2021 | Show parent
  #242
Lives for gear
 
burns46824's Avatar
 
1 Review written
🎧 10 years
Quote:
Originally Posted by Snoggin ➡️
If anyone has any interest here’s my first mix with my 1608 full of red dots. Straight off 16 track tape machine. 1608 2 buss into pro tools.. all analog processing no editing or punches

That sounds quite good, considering that Soundcloud really lowers the resolution of the music...nice work! One thing...did you do a phase check of the overheads and snare? Something is up with those elements. Otherwise...sounding good!
Old 1st March 2021 | Show parent
  #243
Lives for gear
 
🎧 10 years
Quote:
Originally Posted by burns46824 ➡️
That sounds quite good, considering that Soundcloud really lowers the resolution of the music...nice work! One thing...did you do a phase check of the overheads and snare? Something is up with those elements. Otherwise...sounding good!
Hey there Burns! Yes we did check and we hear what you are talking about. When we tracked originally just the drums and acoustic guitar I also recorded into pro tools and I did a wave form check on those elements.

Looked good to us at the time.. I summed the snare top and bottom mics to one Buss on the board and tape with just a bit of bottom snare in the mix... So not sure where we went wrong but Im trying to learn all the time

Any suggestions?

Thanks Tim
Old 1st March 2021 | Show parent
  #244
Lives for gear
 
burns46824's Avatar
 
1 Review written
🎧 10 years
Quote:
Originally Posted by Snoggin ➡️
Hey there Burns! Yes we did check and we hear what you are talking about.

Any suggestions?
I'm actually the least knowledgable person on this probably on the entire thread, because I do synth pop and therefore use LinnDrums and 808s, LOL. Basically, what it's sounding like is the snare is in full stereo...like it's taking up the entire track. Obviously, you don't want that. Perhaps someone with drum recording experience can chime in...?

What 16-track was that recorded on?
Old 1st March 2021 | Show parent
  #245
Lives for gear
 
🎧 10 years
Quote:
Originally Posted by burns46824 ➡️
I'm actually the least knowledgable person on this probably on the entire thread, because I do synth pop and therefore use LinnDrums and 808s, LOL. Basically, what it's sounding like is the snare is in full stereo...like it's taking up the entire track. Obviously, you don't want that. Perhaps someone with drum recording experience can chime in...?

What 16-track was that recorded on?
MCI JH-24 Mara machine I got a few years ago. Thanks.. To us it almost sounds like a flam happening .. Jan the drummer thinks it might be in the tuning of the snare.

Vowed to do better on the next one! Trying hard to get good sounds on the way in
T
Old 4 weeks ago
  #246
Lives for gear
 
burns46824's Avatar
 
1 Review written
🎧 10 years
Zombie Rec., a Swiss designer of op amps, was kind enough to send me 2 different discrete op amp designs that they do (the 5058G and the 3043+) to test against the API 2520s and the Red Dots. In this test, I put them in an API 2500 compressor instead of in my API console. So this is a mix running through the 2500, with each of the op amp sets.

https://www.dropbox.com/sh/ylbr56aan...25NjXwsYa?dl=0
Old 4 weeks ago | Show parent
  #247
Lives for gear
 
IanBSC's Avatar
 
🎧 5 years
Quote:
Originally Posted by burns46824 ➡️
Zombie Rec., a Swiss designer of op amps, was kind enough to send me 2 different discrete op amp designs that they do (the 5058G and the 3043+) to test against the API 2520s and the Red Dots. In this test, I put them in an API 2500 compressor instead of in my API console. So this is a mix running through the 2500, with each of the op amp sets.

https://www.dropbox.com/sh/ylbr56aan...25NjXwsYa?dl=0
I'm always a fan of the Red Dots, but it seemed a tad soft here. The 5058G and 3043+ were nice, but just didn't feel quite right.
Old 3 weeks ago
  #248
Lives for gear
 
burns46824's Avatar
 
1 Review written
🎧 10 years
Perhaps some others want to chime in regarding the API 2500 op amp test with the Zombie Rec. op amps?
Old 3 weeks ago | Show parent
  #249
Quote:
Originally Posted by burns46824 ➡️
Perhaps some others want to chime in regarding the API 2500 op amp test with the Zombie Rec. op amps?
I have tested the Zombie opamps in a few scenarios and had no complaints. What's weird is that I have not yet managed to do a listening test.

Here's a few screenshots of the distortion performance of some discrete and IC opamps driving the output transformer of a SoundSkulptor EQP501 (the default OPA2134 is in the front). You can see that the Zombie opamp does very well. The RED25 does, too BTW.

Didn't I post these before?

Edit: Take what I said here with a grain of salt. John Mulcahy (the REW creator) has pointed me to stepped sine measurements which produce different results.
Attached Thumbnails
API 1608 Mix Buss - Op Amp Blind Listening Test - API 2520, 990C, Rogue 5, GAR 2520, VF600, Red Dot-5058g.jpg   API 1608 Mix Buss - Op Amp Blind Listening Test - API 2520, 990C, Rogue 5, GAR 2520, VF600, Red Dot-ca0252.jpg   API 1608 Mix Buss - Op Amp Blind Listening Test - API 2520, 990C, Rogue 5, GAR 2520, VF600, Red Dot-gar2520.jpg   API 1608 Mix Buss - Op Amp Blind Listening Test - API 2520, 990C, Rogue 5, GAR 2520, VF600, Red Dot-lme49710.jpg   API 1608 Mix Buss - Op Amp Blind Listening Test - API 2520, 990C, Rogue 5, GAR 2520, VF600, Red Dot-lme49990.jpg  

API 1608 Mix Buss - Op Amp Blind Listening Test - API 2520, 990C, Rogue 5, GAR 2520, VF600, Red Dot-lt1128.jpg   API 1608 Mix Buss - Op Amp Blind Listening Test - API 2520, 990C, Rogue 5, GAR 2520, VF600, Red Dot-oa10.jpg   API 1608 Mix Buss - Op Amp Blind Listening Test - API 2520, 990C, Rogue 5, GAR 2520, VF600, Red Dot-red25.jpg   API 1608 Mix Buss - Op Amp Blind Listening Test - API 2520, 990C, Rogue 5, GAR 2520, VF600, Red Dot-ss3601.jpg  
Old 3 weeks ago | Show parent
  #250
Gear Maniac
 
🎧 10 years
2520

I usually prefer red dots, but the 2520 is my fav on this, it has a pleasing ‘zing’ to it
Old 3 weeks ago
  #251
Lives for gear
 
burns46824's Avatar
 
1 Review written
🎧 10 years
Any other thoughts on the Zombie Rec. op amps in this test?
Old 3 weeks ago | Show parent
  #252
Gear Nut
 
coso's Avatar
 
🎧 5 years
Quote:
Originally Posted by burns46824 ➡️
Perhaps some others want to chime in regarding the API 2500 op amp test with the Zombie Rec. op amps?
I certainly appreciate your effort.

Are the differences between op-amps exaggerated when they are driven hard? For example using high gain at line level through a preamp?

It would be great to hear some less subtle comparison files!

Old 3 weeks ago | Show parent
  #253
Quote:
Originally Posted by coso ➡️
Are the differences between op-amps exaggerated when they are driven hard? For example using high gain at line level through a preamp?
Opamps have different distortion spectra, but I have not checked if it changes as you get closer to the supply voltage limits. Opamps can also "run out of steam" (gain-bandwidth product) at higher gains and roll off highs (or do other weird things). Also, when driving transformers, you sometimes see small irregularities in the frequency response in the discretes, less so in IC opamps.
Old 2 weeks ago | Show parent
  #254
Gear Maniac
 
dogears's Avatar
 
1 Review written
🎧 5 years
Quote:
Originally Posted by johannburkard ➡️
Opamps have different distortion spectra, but I have not checked if it changes as you get closer to the supply voltage limits. Opamps can also "run out of steam" (gain-bandwidth product) at higher gains and roll off highs (or do other weird things). Also, when driving transformers, you sometimes see small irregularities in the frequency response in the discretes, less so in IC opamps.
Most modern op amps use a class AB output stage. Higher current demand levels produce increasing third order distortion in this case with a step change around the limit of the bias current.
Old 2 weeks ago | Show parent
  #255
Quote:
Originally Posted by johannburkard ➡️
I have tested the Zombie opamps in a few scenarios and had no complaints. What's weird is that I have not yet managed to do a listening test.

Here's a few screenshots of the distortion performance of some discrete and IC opamps driving the output transformer of a SoundSkulptor EQP501 (the default OPA2134 is in the front). You can see that the Zombie opamp does very well. The RED25 does, too BTW.
The LME49990 (discontinued) has much better THD performance than those plots. On the Audio Precision it reads the residuals at .0005% THD+noise. Do the noise gain trick and it's much lower. That's how the manufacturers test and quantify these devices.
Old 2 weeks ago | Show parent
  #256
Quote:
Originally Posted by Jim Williams ➡️
The LME49990 (discontinued) has much better THD performance than those plots.
I have measured the LME49990 in the discrete opamp spot of the EQP501 and the distortion performance is really good but that of the DIY.RE RED25, Analog LT1128, Sparkos SS3601 and CAPI OA10 were slightly better there -- roughly 1.5-2 dB at 1 kHz if I can trust REW.

On the other hand, in the EQP501, the 3rd harmonic seems to be pretty much independent of the opamp, making it difficult to see what distortion characteristics an opamp has.

Edit: Take what I said here with a grain of salt. John Mulcahy (the REW creator) has pointed me to stepped sine measurements which produce different results.
Attached Thumbnails
API 1608 Mix Buss - Op Amp Blind Listening Test - API 2520, 990C, Rogue 5, GAR 2520, VF600, Red Dot-red25.png   API 1608 Mix Buss - Op Amp Blind Listening Test - API 2520, 990C, Rogue 5, GAR 2520, VF600, Red Dot-lme49990.png  

Last edited by johannburkard; 2 weeks ago at 08:56 PM.. Reason: New wisdom obtained
Old 2 weeks ago | Show parent
  #257
Quote:
Originally Posted by dogears ➡️
Higher current demand levels produce increasing third order distortion in this case with a step change around the limit of the bias current.
Can you see this behaviour in the screenshots I posted above?
Old 2 weeks ago | Show parent
  #258
Gear Maniac
 
dogears's Avatar
 
1 Review written
🎧 5 years
Quote:
Originally Posted by johannburkard ➡️
Can you see this behaviour in the screenshots I posted above?
Not sure - the way I'd expect to see THD presented would be at a fixed frequency (1k, or 10k if you want to be more demanding) and then an FFT of the output signal with the signal components broken out.

Terminate the output into a 10k ohm load, then take a look at THD at +0 dBu and +20 dBu. Then terminate with a 600 ohm load and repeat.

If you can find where the output stage's bias current is overcome, there will be an increase in 3rd order as it transitions to class AB.
Old 1 week ago | Show parent
  #259
Quote:
Originally Posted by burns46824 ➡️
Any other thoughts on the Zombie Rec. op amps in this test?
Yesterday, I have recorded all kinds of IC and discrete opamps in the Colour Palette and the EQP501.

In the EQP501, I tested the DIY.RE RED25, TI LME49990, Sparkos SS3601, Analog LT1128, CAPI CA0252, Zombie 5058G and CAPI GAR2520. Note this was not a blind test.

It felt to me like the Sparkos and the LT1128 were the cleanest/most transparent in the bass and the CA0252 and GAR2520 the least transparent. The 5058G was maybe somewhere in the middle but the differences were extremely small.

In the blind Colour Palette test, I rated the original lowest and minimally brighter opamps higher, with the Analog ADA4625-2, Orange discrete opamp, Analog AD823, TI OPA1612 and TI OPA1656 being the top choices.
Old 1 week ago
  #260
On a less subjective note, if you're interested in evaluating opamps in a more scientifc way, check out DeltaWave: https://deltaw.org/. I've used ABX test for transparency tests with surprising results and X/Y preference tests in other cases.
Old 1 week ago | Show parent
  #261
Quote:
Originally Posted by johannburkard ➡️
I have measured the LME49990 in the discrete opamp spot of the EQP501 and the distortion performance is really good but that of the DIY.RE RED25, Analog LT1128, Sparkos SS3601 and CAPI OA10 were slightly better there -- roughly 1.5-2 dB at 1 kHz if I can trust REW.

On the other hand, in the EQP501, the 3rd harmonic seems to be pretty much independent of the opamp, making it difficult to see what distortion characteristics an opamp has.

Edit: Take what I said here with a grain of salt. John Mulcahy (the REW creator) has pointed me to stepped sine measurements which produce different results.
Without level references those are meaningless and not near the real THD. The 49990 has THD around -156dbu according to the manufacturer. Only the best test gear (Audio Precision) and noise gain tricks can be used to measure those tiny levels.

I don't take much meaning of sine wave 1k measurements anyway, it's cheating as that is the easiest frequency to reproduce. Try it at 20k hz and see what happens.

I prefer CCIF IMD tests (usually reading 1.5 ppm on a good circuit) or FFT's that show all the harmonics.

Of course the OLG at 10k Hz will tell "the rest of the story". This is where most discretes fall off, exception being the Jensen 990.
Old 1 week ago | Show parent
  #262
Quote:
Originally Posted by Jim Williams ➡️
Without level references those are meaningless and not near the real THD.
I agree with that, but not completely useless if I'm hunting for the lowest distortion I can get with the opamps I have.

Quote:
Originally Posted by Jim Williams ➡️
I don't take much meaning of sine wave 1k measurements anyway, it's cheating as that is the easiest frequency to reproduce.
Yes, and I'm sure chip manufacturers optimize for 1 kHz distortion because most distortion curves dip down substantially in that area.

Quote:
Originally Posted by Jim Williams ➡️
Of course the OLG at 10k Hz will tell "the rest of the story". This is where most discretes fall off, exception being the Jensen 990.
Sometimes, you don't need very much gain.
Old 1 week ago
  #263
@ Jim Williams on discretes, have you tried out any of the "Hifi" discretes? Sparkos, Sonic Imagery Labs, NewClassD, Staccato, Burson, Orange etc.?
Old 1 week ago | Show parent
  #264
Quote:
Originally Posted by johannburkard ➡️
@ Jim Williams on discretes, have you tried out any of the "Hifi" discretes? Sparkos, Sonic Imagery Labs, NewClassD, Staccato, Burson, Orange etc.?
Not many. I tend to prefer a good discrete circuit or a good IC opamp design over discrete opamps. I was familiar with the old AD discretes from the 1960's. IC's allowed for better transistor matching and less parasitics = greater bandwidth, higher open loop gain and less noise and THD.

Many of the current discretes could use a complete test and comparison but no one tends to care about that stuff, they are selected for sound.
Old 1 week ago | Show parent
  #265
Quote:
Originally Posted by Jim Williams ➡️
Many of the current discretes could use a complete test and comparison but no one tends to care about that stuff, they are selected for sound.
Not sure Sparkos and SIL would agree with that but I did a blind test with the Orange discrete:

Code:
Trial 1, user: B actual: B	 1/1
Trial 2, user: A actual: A	 2/2
Trial 3, user: B actual: B	 3/3
Trial 4, user: B actual: B	 4/4
Trial 5, user: B actual: B	 5/5
Trial 6, user: A actual: A	 6/6
Trial 7, user: B actual: B	 7/7

Probability of guessing: 0,8%
A=Reference,B=Comparison
Test type: ABX

-----------------
DeltaWave v1.0.57, 2021-04-26T21:22:12.9198953+02:00
Reference:  colour-direct.wav[L] 671999 samples 44100Hz 16bits, ch=0, MD5=00
Comparison: colour-orange.wav[L] 671999 samples 44100Hz 16bits, ch=0, MD5=00
LastResult: Ref=C:\Users\Johann\Music\Diverses\Opamps\colour-direct.wav:[L], Comp=C:\Users\Johann\Music\Diverses\Opamps\colour-orange.wav:[L]
LastResult: FilterType=0, Bandwidth=0
LastResult: FilterType=0, Bandwidth=0
LastResult: Drift=-4,59441929984332E-09, Offset=-0,6498955866 Phase Inverted=False
LastResult: Freq=44100

Settings: 
	Gain:True, Remove DC:True
	Non-linear Gain EQ:False	Non-linear Phase EQ: False
	EQ FFT Size:65536, EQ Frequency Cut: 0Hz - 0Hz, EQ Threshold: -160dB
	Correct Drift:True, Precision:30
	Non-Linear drift Correction:False
	Upsample:False, Window:Hann
	Spectrum Window:Hann, Spectrum Size:32768
	Spectrogram Window:Hann, Spectrogram Size:4096, Spectrogram Steps:2048
	Dither:False
	Trim Silence:False
	Enable Simple Waveform Measurement: False


Signature: 8994e3470105bf4b4003117b7ea395e6
Not that the Orange sounded bad or anything.
Old 1 week ago
  #266
I'd like to see complete tests done with an AP analyzer. Most of the current IC offerings provide extensive test results, it is informative to those that can interpret them.

One spec not mentioned is slew rate. I don't know of any discrete manufacturer providing those specs other than the 990 at 20V/us.
Old 1 week ago | Show parent
  #267
Gear Maniac
 
dogears's Avatar
 
1 Review written
🎧 5 years
Quote:
Originally Posted by Jim Williams ➡️
I'd like to see complete tests done with an AP analyzer. Most of the current IC offerings provide extensive test results, it is informative to those that can interpret them.

One spec not mentioned is slew rate. I don't know of any discrete manufacturer providing those specs other than the 990 at 20V/us.
https://www.ironageaudioworks.com/m2doa

Old 1 week ago | Show parent
  #268
Quote:
Originally Posted by Jim Williams ➡️
I don't know of any discrete manufacturer providing those specs other than the 990 at 20V/us.
Orange says 52 V/µs typical. Sonic Imagery Labs 18-19 V/µs.
Old 1 week ago | Show parent
  #269
Gear Maniac
 
dogears's Avatar
 
1 Review written
🎧 5 years
Quote:
Originally Posted by johannburkard ➡️
Orange says 52 V/µs typical. Sonic Imagery Labs 18-19 V/µs.
Honestly I have no idea what the practical benefit is to have these very high slew rates. They seem to me like marketing.

Practically speaking audio is bounded by modern voltage levels and the audible spectrum. Even if we're extremely generous on both of these you don't have a need for monstrous slew rates.
Old 1 week ago | Show parent
  #270
Quote:
Originally Posted by dogears ➡️
Honestly I have no idea what the practical benefit is to have these very high slew rates. They seem to me like marketing.
Posting for reference,

Quote:
On record ticks and pops I could register slew rates as high as 20–30V/µs [...] In playing music we were looking at piano, cymbals, violins—a large range of instruments all played loud, near the power capacity of the amplifier. And we were getting values like 0.1V/µs, 0.5V/µs. In fact, at that time in that series of tests, the highest transient we were able to achieve was about 1V/µs. So the 10:1 ratio would dictate that you would need at least 10V/µs performance. [...] Subsequent to that, though, the Sheffield Drum Record was used in a test and we started getting figures into the 5V/µs ballpark. [...] But given that you're on the 5V/µs rate, we ultimately settled on approximately 50V/µs as a speed beyond which was fine, but it became a minimal figure that we deal with in our designs.
From https://www.stereophile.com/content/...-better-page-4.
📝 Reply

Similar Threads

Thread / Thread Starter Replies / Views Last Post
replies: 87 views: 43047
Avatar for 12tone
12tone 4th May 2017
replies: 295 views: 71000
Avatar for anguswoodhead
anguswoodhead 26th March 2013
replies: 1296 views: 176195
Avatar for heraldo_jones
heraldo_jones 1st February 2016
replies: 28186 views: 2964070
Avatar for RJHollins
RJHollins 1 minute ago
Post Reply

Welcome to the Gearspace Pro Audio Community!

Registration benefits include:
  • The ability to reply to and create new discussions
  • Access to members-only giveaways & competitions
  • Interact with VIP industry experts in our guest Q&As
  • Access to members-only sub forum discussions
  • Access to members-only Chat Room
  • Get INSTANT ACCESS to the world's best private pro audio Classifieds for only USD $20/year
  • Promote your eBay auctions and Reverb.com listings for free
  • Remove this message!
You need an account to post a reply. Create a username and password below and an account will be created and your post entered.


 
 
Slide to join now Processing…

Forum Jump
Forum Jump