Post by: (Brian) Frost on November 13, 2006, 03:33:15 PM
Frost
Post by: Josh Billings on November 13, 2006, 10:31:39 PM
Hope this helps
-Josh Billings
Post by: Ed Walters on November 14, 2006, 06:29:50 AM
I've used/listened to the PL2, PLX, and PLX2 lines extensively.
One consideration when comparing them is to compare similar topologies -- some models are class AB, most are class H; there are subtle differences in sound which do not indicate a difference between lines.
That said, my impression is that the PL2 line is somewhat more neutral and somewhat tighter sounding. They give the impression of having less bottom, when in fact the "less" is less overhang. The PL2 line is very revealing of source and speaker problems. The PLX and PLX2 lines, on the other hand, could be said to be "musical" -- or veiled -- depending on the spin you want to hear/use/sell. These differences are not large -- when we compared them we did it in a reasonable acoustic environment with studio monitors and measured identical amplifier outputs.
If you are going to overdrive the amps (below 2 ohms or well into clipping, etc.) the PL2 will be slightly more robust/recover from overload slightly more cleanly. All of them will actually drive 2 ohm loads with full power test signals long enough to measure the output, unlike some other highly regarded amplifers.
At their price point, the PLX/PLX2 are hard to beat for sound/power/features/reliability. At their price point, the PL2 have a fair amount of competition.
As an aside, if your needs fit the product, the original PL series (what remains of it) are seriously good as well -- but they are deeper, taller, heavier, and can have significantly more power....
Ed Walters
Post by: Bob Lee (QSC) on November 14, 2006, 02:49:22 PM
Post by: jeffhtg (Jeff Kenney) on November 18, 2006, 08:50:05 PM
Both of those amps generate a bit of heat
- but good solid amps!
Post by: Ryan Garnett on November 18, 2006, 10:54:54 PM
Post by: Ken Freeman on November 18, 2006, 10:58:22 PM
Ken
Post by: Bob Leonard on November 19, 2006, 12:35:34 AM
Post by: jeffhtg (Jeff Kenney) on November 19, 2006, 03:07:58 AM
I will say this.. they WILL FAIL. but I have some big mx amps that were installed early 90's and just now need to be recapped or replaced
and THATS in a nightclub with fog , crappy atmospheric control, and running close to clip all the time!Post by: David Gardner on November 19, 2006, 01:08:09 PM
Post by: mel taylor on November 19, 2006, 01:35:40 PM
no doubt that qsc's are reliable and big bang for the buck, but the fidelity is just not there.
you get what you pay for.
Mel
Post by: Andy Peters on November 19, 2006, 02:09:08 PM
| mel taylor wrote on Sun, 19 November 2006 11:35 |
For those of you who are concerned w/ (can hear) sound quality, try a blind comparison w/ any qsc (or new generation crown) vs. any professional concert amp... lab g, camco, dynacord, etc. no doubt that qsc's (sic) are reliable and big bang for the buck, but the fidelity is just not there. you get what you pay for. |
Please describe your test set-up. Show all work.
-a
Post by: John Horvath on November 19, 2006, 02:09:09 PM
| mel taylor wrote on Sun, 19 November 2006 12:35 |
try a blind comparison w/ any qsc (or new generation crown) vs. any professional concert amp... lab g, camco, dynacord, etc. no doubt that qsc's are reliable and big bang for the buck, but the fidelity is just not there. |
Oh great, here we go again..
Post by: mel taylor on November 19, 2006, 04:30:01 PM
| Andy Peters wrote on Sun, 19 November 2006 19:09 |
Please describe your test set-up. Show all work. -a |
venice output into DN 1248+, into various amps, into passive switcher, into one 8Ohm-two way cab. Match sensitivities by attenuators, listen and switch.
very simple.
mt
Post by: Ken Freeman on November 19, 2006, 05:24:02 PM
Hey Mel, not buying into the description of your critical listening test. You are going to have to describe your double blind listening test and why you used a speaker processor that I regulary bypass because it sounds like crap. I have hunderds of amplifiers from the likes of Crest, QSC, Crown, BGW, Ramsa, Apogee, UREI, EV, Yamaha, etc. We bench these all the time and rarely find a big difference between any of the majors brands other than output power and how they behave under stress. You are entitled to your opinion, of course, but understand that you will be challenged on your findings when you throw really good products under the bus...
Ken
Post by: Geri O'Neil on November 19, 2006, 06:45:19 PM
| mel taylor wrote on Sun, 19 November 2006 12:35 |
For those of you who are concerned w/ (can hear) sound quality, try a blind comparison w/ any qsc (or new generation crown) vs. any professional concert amp... lab g, camco, dynacord, etc. no doubt that qsc's are reliable and big bang for the buck, but the fidelity is just not there. you get what you pay for. Mel |
Mel, yours is an opinion that will be shared with next to no one. We have done blind listening tests with the Pls, PLX series and Labs among others. While we are a Lab Gruppen house, we could find no audible differences that we could verify or consistently reproduce.
Oh, and by the way, your condescending statement of "those whoe are concerned/can hear sound quality" wreaks of holier-than-thou"-ism. It means nothing in a professional forum. Why can't you just offer an opinion, popular or otherwise, without some little wordbombs such as that? It gets you nowhere.
Yes, you are entitled you your opinion, yada yada ad naseum beyond belief-eum (yes, a little gem from the GSS manual)
Geri "I kinda like this contrarian thing, even though I don't do it so well" O
Post by: Scott Flaws on November 19, 2006, 07:09:08 PM
I have over 30 QSC amps in my invetory and wouldnt trade them for anything else.
Post by: Ryan Lantzy on November 20, 2006, 12:15:36 AM
| mel taylor wrote on Sun, 19 November 2006 13:35 |
For those of you who are concerned w/ (can hear) sound quality, try a blind comparison w/ any qsc (or new generation crown) vs. any professional concert amp... lab g, camco, dynacord, etc. no doubt that qsc's are reliable and big bang for the buck, but the fidelity is just not there. |
Bull*cough*shurgggt*cough*
Post by: mel taylor on November 20, 2006, 09:43:11 AM
| Ken Freeman wrote on Sun, 19 November 2006 22:24 |
Yeah and I like my 40 year old Mac MC 240 way better than all this newfangled transistor stuff....All 40 watts of it! Hey Mel, not buying into the description of your critical listening test. You are going to have to describe your double blind listening test and why you used a speaker processor that I regulary bypass because it sounds like crap. I have hunderds of amplifiers from the likes of Crest, QSC, Crown, BGW, Ramsa, Apogee, UREI, EV, Yamaha, etc. We bench these all the time and rarely find a big difference between any of the majors brands other than output power and how they behave under stress. You are entitled to your opinion, of course, but understand that you will be challenged on your findings when you throw really good products under the bus... Ken |
Ken,
no speaker processor used in my tests.
You are also entitled to your opinion, and from your expressed opinions here, I likely wouldn't get along with you when it comes to audio, no offense personally.
I'm not throwing any products under the bus. Their amps are good, reliable, lot's o' watts to $s, etc. Just like other companies, they have their place. For me, it's just not when you demand the best.
Post by: mel taylor on November 20, 2006, 09:57:28 AM
| Geri O wrote on Sun, 19 November 2006 23:45 |
Mel, yours is an opinion that will be shared with next to no one. We have done blind listening tests with the Pls, PLX series and Labs among others. While we are a Lab Gruppen house, we could find no audible differences that we could verify or consistently reproduce. Oh, and by the way, your condescending statement of "those whoe are concerned/can hear sound quality" wreaks of holier-than-thou"-ism. It means nothing in a professional forum. Why can't you just offer an opinion, popular or otherwise, without some little wordbombs such as that? It gets you nowhere. Yes, you are entitled you your opinion, yada yada ad naseum beyond belief-eum (yes, a little gem from the GSS manual) |
Geri,
No problem if you don't agree. I don't need yours, or anyone on the LABs reinforcement to believe what I know. I have plenty of trusted colleagues in the industry whom I collaborate w/ and question for second opinions. Let's just say we disagree.
I didn't intend to offend anyone, so please don't be so defensive. Not every job requires the best fidelity possible, and this should be considered when specifying equipment- - that was all I meant. However, again, for those engineers and applications requiring the highest possible fidelity at every point in the signal chain, there is no room for "good enough".
Post by: John Roberts {JR} on November 20, 2006, 10:45:54 AM
| mel taylor wrote on Sun, 19 November 2006 12:35 |
For those of you who are concerned w/ (can hear) sound quality, try a blind comparison w/ any qsc (or new generation crown) vs. any professional concert amp... lab g, camco, dynacord, etc. no doubt that qsc's are reliable and big bang for the buck, but the fidelity is just not there. you get what you pay for. Mel |
Most sound professionals knowledgeable about power amps will not be comfortable with your claim that "sound quality" differences are audible in simple listening tests. When listening to a signal chain composed of multiple blocks some of which exhibit much higher levels of nonlinearity, such tests can be compromised by the other weaker links.
You may be able to hear differences in amplifier design philosophy wrt to how they handle; clipping (limiters, PS capacitance, etc), minimum load impedances (current and VI limiting), and even design/manufacturing flaws (excessive crossover or class G-H transition perturbations).
Sound quality, brings to mind characteristics like linearity (THD and IMD), frequency response, and S/N which will be more similar than different between quality professional amplifiers and difficult to parse out in even double blind listening tests (well maybe not noise floor when the music stops).
Perceived differences will more often be caused by simple things like rated or available power. Some designs (like PFC) will be better at extracting power from a given distro so may sound different when pushed hard. There is a place for listening tests to compare how amplifiers act when pushed beyond their rated capabilities, but lets not confuse that with sound quality unless you consider that normal operation.
FWIW amp designers invest a great deal of effort to overload as benignly as possible. Such attention to detail is an aspect of product quality, but more like air bags in a car than a sport handling package (sorry to revert to yet another automotive analogy).
JR
PS: I'm proud I avoided the obvious jokes (viewing art through sun glasses, tasting fine wine in a dirty glass, or that scatological one about judging a fine restaurant meal by.... I'll let you all digest that last one.
Post by: Bob Lee (QSC) on November 20, 2006, 12:04:33 PM
| mel taylor wrote on Sun, 19 November 2006 10:35 |
For those of you who are concerned w/ (can hear) sound quality, try a blind comparison w/ any qsc (or new generation crown) vs. any professional concert amp... lab g, camco, dynacord, etc. no doubt that qsc's are reliable and big bang for the buck, but the fidelity is just not there. you get what you pay for. Mel |
That's funny … the fidelity was there when they left the factory …
Post by: Marty Bilecki on November 20, 2006, 03:01:55 PM
Did UPS drop the box and the fidelity was damaged in shipping??
Marty B
Post by: Scott Flaws on November 20, 2006, 03:06:27 PM
Post by: Marty Bilecki on November 20, 2006, 03:11:03 PM
And also funny that my company owns 21 Plx series amps and 6 Rmx 5050's and not one has ever left the rack it was installed in !!!!1
I just got back from service I tech amp number 7 of the 11 I techs I have owned for the last 2 years..
(Reading from the repair invoice in my pocket
" Filter Cap on Usp3t shorted, damage went back to power supply. Replaced Ups3t and input boards , repaired low voltage power supply. Adjusted and tested amplifier to factory specifications. Restored firmware to version 2.048dsp and 2.031Disp. and reloaded customer preset data..")
I must say the turnaround time has been great on all of the amps that have pooped out.. I sent this amp back to Crown on Nov 13 th and it came back today the 20 th..
7 out of 11 of my Hi Fidelity Amps !!!!!! Cool Eh?
Marty Bilecki
Post by: Marty Bilecki on November 20, 2006, 03:15:04 PM
mb
Post by: Adam Kane on November 20, 2006, 05:21:59 PM
Looks like you've got one great set of ears on ya. From my experience, the only times I've ever been able to hear differences between amps (even with cheapies) was when they were driven into oblivion...something I try hard to prevent by properly sizing the amp to the application and the system for the job. I believe JR already mentioned something along these lines so I'll go no further.
To the OP,
I can't remember the last time I've had a QSC product fail. They sound great (unless the fidelity leaks out of the box during shipping) and seem to be bullet-proof. We sell, install, and use their amps with confidence. Every time one leaves the shop, we just know that there won't be problems. I can't think of many (if any) other brands I can say that about.
Post by: mel taylor on November 20, 2006, 05:58:31 PM
| Adam Kane wrote on Mon, 20 November 2006 22:21 |
Mel, Looks like you've got one great set of ears on ya. From my experience, the only times I've ever been able to hear differences between amps (even with cheapies) was when they were driven into oblivion...something I try hard to prevent by properly sizing the amp to the application and the system for the job. I believe JR already mentioned something along these lines so I'll go no further. |
| mel taylor wrote on Mon, 20 November 2006 14:43 |
I'm not throwing any products under the bus. Their amps are good, reliable, lot's o' watts to $s, etc. Just like other companies, they have their place. For me, it's just not when you demand the best. |
Again, no doubt that qsc is a leader in value and reliability, no argument.
Of course most don't hear the difference, consider "most" of the concerts, or otherwise amplified audio events all of us have heard over the years. "Most" come with so much distortion and other influencing factors that the fidelity of an amplifier cannot be distinguished by any human. OTOH, many ears in the business (several I know in Michigan Adam),are always striving for the next level of fidelity. To my ears, the perfect sound system would not be heard. I believe with a professional system that's properly optimized, the nuance of sound quality can be distinguished in amplifiers, as well as consoles, etc. Of course the difference may not be realized if the amplifier is not near the weakest link in the signal chain. However, it makes a difference when you get everything as close to perfect as possible. This is the dramatic difference in "pristine" vs. "good" sound quality that "most" cannot respect, as they've never experienced it. This is just my opinion, and isn't meant to change anyones mind, but I am part of the audio engineering community, and I have many peers that think likewise. .02
Post by: Bob Lee (QSC) on November 20, 2006, 06:05:53 PM
Have you ever tried a double-blind test?
And I'm serious about the fidelity being in the amps at the factory; the PLX and PowerLight amps in particular would make great studio reference amps if it weren't for their fans. Do you know what went wrong with the ones you tried?
Post by: mel taylor on November 20, 2006, 06:29:03 PM
| Bob Lee (QSC) wrote on Mon, 20 November 2006 23:05 |
You are not alone, Mel. I've heard the same sentiment expressed by guys at Guitar Center and the now-defunct Mars Music, but OTOH I don't think those guys know how to compare power amps. Have you ever tried a double-blind test? And I'm serious about the fidelity being in the amps at the factory; the PLX and PowerLight amps in particular would make great studio reference amps if it weren't for their fans. Do you know what went wrong with the ones you tried? |
Thanks Bob,
I don't shop @ GC, and I'm serious too.
I am speaking of double-blind tests, many times and places including, outdoors, in a theatre, and more than once in a recording studio.
The PLX are normally, among others, one of the first to go.
I've nearly never experienced a qsc failure or fault. One did smoke quickly during 2 Ohm testing - an infant failure. Rarely, if ever, during an event or install.
Post by: Bob Lee (QSC) on November 20, 2006, 06:55:54 PM
So, were they real double-blind tests? Many people don't know what the correct methodology is. The PLX and PowerLight amps will really reproduce faithfully what you put into them, up until the point of clipping, so maybe what really displeased you was the input signal.
Post by: Iain_Macdonald on November 20, 2006, 08:15:40 PM
Everything that JR says plus: the load[speaker] needs to be taken in to consideration. Studio speakers are just as bad as many PA products for presenting 'impossible' loads. I do agree with you that some amps sound better than others. Just try one of the large Chord amps and hear the difference it makes. But I do sound a note of caution. The human hearing system is adaptive, and it can adapt very quickly to what you hear. Any subsequent listening in the very short term can be colored by your immediate past experience. That is why double blind test methodology is preferred. I would also agree with your point on the degradation of the signal chain. The system is only as good as the weakest link. So many products are compromised, it's often hard to get a true picture of what's happening. This is especially true of digital products. Though the Lake Contour is a shining example of a clearly better product.
But I would say that we need to be pragmatists, however unpalatable that may be to you.
Iain.
Post by: Tim Duffin on November 21, 2006, 08:38:44 PM
I can't believe this sentence was actually written by an amplifier engineer:
"The PLX and PowerLight amps will really reproduce faithfully what you put into them, up until the point of clipping, so maybe what really displeased you was the input signal."
Whatever happened to frequency response, total harmonic distortion, intermodulation distortion, crossover distortion, etc? No audio device capable of reproducting sound has 0 distortion. Adding qualifiers to statements like "only when used properly" or "out of clipping" says nothing about what happens to an amp when it is ACTUALLY USED. I too could design an amp that has .000000001% THD if it is only required to source 1/400th of its total power. What really matters is what an amplifier does when it is pushed to its absolute limits. For instance, if it is a BASS amp-- it should be run at 100% RMS and 100% duty cycle inside a rack with sagging electrical mains (gotta factor in the 300KW lighting rig on the same service) in 100 degree sunlight in a black amp rack with heavily compressed 50Hz sine waves for 8 hours.
Then we will know what an amp does after its not on the testbench at the manufacturer under the absolute most favorable conditions that the manufacturer could possibly create.
T
Post by: Don Boomer on November 21, 2006, 08:53:43 PM
See the qualifier here is "up to the point of clipping". That usually means less than 10% of it's rated power. If you run an amp with proper headroom you're just not gonna "hear" the amp. What you "hear" is how the amp handles overloads, how it recovers, etc. ABX tests time after time prove that out ... and that's under critical listening conditions. Live sound rarely gets close.
How the amp is actually used I guess depends on if the guy knows how to use it
Post by: Doug Fowler on November 21, 2006, 09:15:29 PM
| Quote: |
Whatever happened to frequency response, total harmonic distortion, intermodulation distortion, crossover distortion, etc? No audio device capable of reproducting sound has 0 distortion. Adding qualifiers to statements like "only when used properly" or "out of clipping" says nothing about what happens to an amp when it is ACTUALLY USED. I too could design an amp that has .000000001% THD if it is only required to source 1/400th of its total power. What really matters is what an amplifier does when it is pushed to its absolute limits. For instance, if it is a BASS amp-- it should be run at 100% RMS and 100% duty cycle inside a rack with sagging electrical mains (gotta factor in the 300KW lighting rig on the same service) in 100 degree sunlight in a black amp rack with heavily compressed 50Hz sine waves for 8 hours. |
All this from a guy who thinks MacroTechs suck and who doesn't meter his power at gigs.
The chance of you outliving this stigma are about as slim as Miffe getting away from that photo JJJ likes to share. Me, I think it's a damn fine photo and it sort of reminds me of my senior photo in the old high school yearbook.
But that was a long time ago, of course.
Post by: Tim Padrick on November 21, 2006, 09:20:09 PM
Linn Klout
Arcam P90
Many steps below those:
Crown K1 (tested in both stereo and bridged).
Several steps below that:
Crown K2.
Very close behind were the Crown Micro 1200 and CE4000.
A step or two behind those was the Mackie 1400.
Of course the first two are of such low power and duty cycle capacility that they are useful only for home hi-fi and perhaps studio monitoring purposes. Ignoring them, the K1 was quite a bit better than the others. (I also compared the K1 to a K2 in a bass rig, and again the K1 sounded a lot better when not driven to clipping.) The K1 of course is only enough power for horns and maybe wedges that are very efficient or not driven very loud (unless the K1 is bridged - in which mode it still sounds better than the others even when driven to its limits).
Post by: John Roberts {JR} on November 21, 2006, 09:36:20 PM
| Tim Duffin wrote on Tue, 21 November 2006 19:38 |
Bob... I can't believe this sentence was actually written by an amplifier engineer: "The PLX and PowerLight amps will really reproduce faithfully what you put into them, up until the point of clipping, so maybe what really displeased you was the input signal." Whatever happened to frequency response, total harmonic distortion, intermodulation distortion, crossover distortion, etc? No audio device capable of reproducting sound has 0 distortion. Adding qualifiers to statements like "only when used properly" or "out of clipping" says nothing about what happens to an amp when it is ACTUALLY USED. I too could design an amp that has .000000001% THD if it is only required to source 1/400th of its total power. What really matters is what an amplifier does when it is pushed to its absolute limits. For instance, if it is a BASS amp-- it should be run at 100% RMS and 100% duty cycle inside a rack with sagging electrical mains (gotta factor in the 300KW lighting rig on the same service) in 100 degree sunlight in a black amp rack with heavily compressed 50Hz sine waves for 8 hours. Then we will know what an amp does after its not on the testbench at the manufacturer under the absolute most favorable conditions that the manufacturer could possibly create. T |
For instance, if it is a BASS amp-- it should be run at 100% RMS and 100% duty cycle inside a rack with sagging electrical mains (gotta factor in the 300KW lighting rig on the same service) in 100 degree sunlight in a black amp rack with heavily compressed 50Hz sine waves for 8 hours.
100% RMS ?? Do you perhaps mean 100% rated power? At which impedance? Which rated power?
100% duty cycle.. OK full on, for all the time...
Inside a rack ? OK.
With sagging electrical mains... Um.. that could be a problem. A typical amp won't put out full rated power if it doesn't have rated mains voltage available. How an amp will act with or w/o a clip limiter turned on will make a huge difference in that situation (clipping).
100' sunlight. 100' ambient, with a hot rack apparently...
BTW, how do you compress a 50 Hz sine wave?? If it was already running at full power applying heavy compression will just turn it down??
8 hours... Now that I understand... 480 minutes...
==============
I'm afraid I can't make much sense of your comments. If your point is that amplifiers clip and operate in their nonlinear region when driven hard, I'm inclined to agree.
Your ranting makes any real argument extremely difficult to parse out and therefore it is very easy to dismiss the whole screed. If you have a specific point please make it more clearly.
JR
PS: FWIW I don't know of too many amps that will put out full rated power for 8 hrs continuously.
Post by: Tim Duffin on November 21, 2006, 10:43:22 PM
T
Post by: Tim Duffin on November 21, 2006, 11:00:25 PM
| John Roberts {JR} wrote on Wed, 22 November 2006 02:36 | ||
For instance, if it is a BASS amp-- it should be run at 100% RMS and 100% duty cycle inside a rack with sagging electrical mains (gotta factor in the 300KW lighting rig on the same service) in 100 degree sunlight in a black amp rack with heavily compressed 50Hz sine waves for 8 hours. 100% RMS ?? Do you perhaps mean 100% rated power? At which impedance? Which rated power? 100% RMS is what the manufacturer states is the amps RMS--at whichever impedance the manufacturer states GIVES the HIGHEST rated power--and finally the AES standard power rating conventionally used today. (Do I really have to type this, I'm assuming that people reading this have a certain degree of experience in test engineering or at least reading the owners manual) 100% duty cycle.. OK full on, for all the time... Inside a rack ? OK. With sagging electrical mains... Um.. that could be a problem. A typical amp won't put out full rated power if it doesn't have rated mains voltage available. How an amp will act with or w/o a clip limiter turned on will make a huge difference in that situation (clipping). Really... I thought that multi rail amplifiers tended to deal with sagging mains more efficiently than dual rail amps. Sagging mains are a reality-- so why not rate a particular amps performance with that in mind? 100' sunlight. 100' ambient, with a hot rack apparently... BTW, how do you compress a 50 Hz sine wave?? If it was already running at full power applying heavy compression will just turn it down?? Oops, that was a typo-- I was thinking faster than I was typing. I meant either heavily compressed continuous LF or a 50hz sine wave. If a sine wave generator is not available, just play some jungle or bass music to test an amps LF performance--not Van Halen. Incidentally, I agree that there aren't any amps that put out their "rated" power after 8 hours-- I want to know what they do put out after 8 hours straight and call that the "I am actually using this amp in a situation which is not in a lab so what does it actually put out" power. Then, we can compare the "rated" power to the "real" power specs and see which amps live up to their ratings. 8 hours... Now that I understand... 480 minutes... ============== I'm afraid I can't make much sense of your comments. If your point is that amplifiers clip and operate in their nonlinear region when driven hard, I'm inclined to agree. Your ranting makes any real argument extremely difficult to parse out and therefore it is very easy to dismiss the whole screed. If you have a specific point please make it more clearly. JR PS: FWIW I don't know of too many amps that will put out full rated power for 8 hrs continuously. |
Post by: Andy Peters on November 22, 2006, 12:47:04 AM
| Doug Fowler wrote on Tue, 21 November 2006 19:15 |
The chance of you outliving this stigma are about as slim as Miffe getting away from that photo JJJ likes to share. Me, I think it's a damn fine photo and it sort of reminds me of my senior photo in the old high school yearbook. |
Dunno where JJJ is right now, so I'll "help" :
Post by: John Roberts {JR} on November 22, 2006, 09:22:49 AM
| Tim Duffin wrote on Tue, 21 November 2006 22:00 |
[ Really... I thought that multi rail amplifiers tended to deal with sagging mains more efficiently than dual rail amps. Sagging mains are a reality-- so why not rate a particular amps performance with that in mind? |
Multi-rail amps are more efficient and will therefore cause less mains sag for a given amount of power delivered to the loudspeakers. PFC power supplies do even better by spreading the current draw over the entire waveform.
I am not aware of any good way to characterize this. Perhaps peak mains current would help a few and confuse the rest.
I would like to see some kind of duty cycle (thermal headroom) quantification that end users could relate to musical genres to predict performance in an application but after years haven't figured out how to do it in such a way that would be embraced by the industry. Until then lighter duty and heavier duty power amps will find their niches in the market by simple experience. Cost impact will drive the size of these market segments.
JR
Post by: mel taylor on November 22, 2006, 09:24:22 AM
| John Roberts {JR} wrote on Mon, 20 November 2006 15:45 |
Most sound professionals knowledgeable about power amps will not be comfortable with your claim that "sound quality" differences are audible in simple listening tests. When listening to a signal chain composed of multiple blocks some of which exhibit much higher levels of nonlinearity, such tests can be compromised by the other weaker links. |
Your point is exactly what I mean JR. However, I strongly disagree w/ your haphazard estimation that most wouldn't agree w/ me. I believe more audio professionals (not posers, but pros I would trust my gig w/) than most here are comfortable to admit would acknowledge the audible characteristic differences in cheaper amps vs. high end stuff. Of course, when the amp is the weakest link. What piece in the electrical signal chain of a Venice, and an amp, would be the least faithful? That's what keeps most people listening ignorantly, I believe. You can't hear the characteristics thru a poorly tuned XYZ rig (insert anything sold on almost any retail floor or catalogue).
| John Roberts {JR} wrote on Mon, 20 November 2006 15:45 |
You may be able to hear differences in amplifier design philosophy wrt to how they handle; clipping (limiters, PS capacitance, etc), minimum load impedances (current and VI limiting), and even design/manufacturing flaws (excessive crossover or class G-H transition perturbations). |
Absolutely, these are the nuances we test after concluding the amplifiers sound quality meets or exceeds expectations, and even a better sounding amp better behave well under severe stress with absolute reliability, or it's not usable either.
| John Roberts {JR} wrote on Mon, 20 November 2006 15:45 |
Sound quality, brings to mind characteristics like linearity (THD and IMD), frequency response, and S/N which will be more similar than different between quality professional amplifiers and difficult to parse out in even double blind listening tests (well maybe not noise floor when the music stops). |
Far from reality JR, ever compare DIM 30(40, 50) tests on a range of pro amps? It's surprising what "pro" amps are terrible at this, or have poorly designed protection circuits that interfere.
If this is difficult to hear, even in a blind test, please input pink noise into as many amps as you like, match the outputs, and compare the noise thru a common speaker. Why is there such a difference in the balance of the noise across amplifiers?
| John Roberts {JR} wrote on Mon, 20 November 2006 15:45 |
Perceived differences will more often be caused by simple things like rated or available power. Some designs (like PFC) will be better at extracting power from a given distro so may sound different when pushed hard. There is a place for listening tests to compare how amplifiers act when pushed beyond their rated capabilities, but lets not confuse that with sound quality unless you consider that normal operation. |
For open sound quality strictly, all tests should be performed well before clipping.
| John Roberts {JR} wrote on Mon, 20 November 2006 15:45 |
FWIW amp designers invest a great deal of effort to overload as benignly as possible. Such attention to detail is an aspect of product quality, but more like air bags in a car than a sport handling package (sorry to revert to yet another automotive analogy). |
Okay, somewhat true. However, the reality is some amps use higher performance, more expensive components inside - from power supply to output transistors. Some don't, so when you compare the schematics of a cheaper "GC ready" amp to something more "unique", even though the power ratings may be same/similar, I will always trust "you get what you pay for".
Post by: Doug Fowler on November 22, 2006, 11:23:19 AM
Post by: John Roberts {JR} on November 22, 2006, 11:30:01 AM
| mel taylor wrote on Wed, 22 November 2006 08:24 |
Far from reality JR, ever compare DIM 30(40, 50) tests on a range of pro amps? It's surprising what "pro" amps are terrible at this, or have poorly designed protection circuits that interfere. If this is difficult to hear, even in a blind test, please input pink noise into as many amps as you like, match the outputs, and compare the noise thru a common speaker. Why is there such a difference in the balance of the noise across amplifiers? |
I'm too busy to give you a full point by point but I will address this one point.
The slew rate wars ran their course several decades ago and (was it Matti Ottola) the champion of those early unconventional tests to parse out slew induced distortions was not aware of the seminal work published in a brief AES journal paper by Marshall Leach (EE prof at Ga tech) that they all can be readily engineered out.
The sound marketplace is conservative to a fault and clings to specification they think they understand. Amplifier slew rate is one such specification. IMO the correct way to quantify a power amplifiers HF performance is with Power bandwidth and rise time.
In a properly designed power amp (or any circuit block for that matter) you should LPF the very input stage such that you cannot exceed the amplifier's open loop slew rate with any valid (unclipped) waveform. In such properly band limited circuits you do not have a meaningful max slew rate but instead a rise-time spec (defined as the time it takes a square wave to traverse from 10% to 90% of it's amplitude). In these properly designed amplifiers a 1V, a .1V, and a 10V (100V if not clipped) square wave all pass cleanly and look "exactly" the same except for amplitude.
Consumers love slew rate specs and consumers get what they want, while it has been a non issue in proper amp design for decades. FWIW I have seen some horrible circuit designs to game the slew rate spec to some silly high number while compromising linearity elsewhere. These approaches have merit in nonlinear "Sample and hold" circuits, but not for audio.
Note: I am not saying all amplifiers are properly designed but most of the majors have the basics very well figured out. There will always be substandard crap if you dig low enough into the bottom of the barrel.
JR
PS: I don't have a good guess at what you're hearing in Pink noise... I surely wouldn't expect slew related factors, perhaps simple frequency response?
Post by: Dave Miller on November 22, 2006, 06:27:16 PM
I would agree that under normal conditions, sonic differences between different amps are minimal and frequently inaudible. The weak links in the chain are your mics and speakers and these introduce more coloration and distortion than any of the electronics. Over the years I've seen amps go in and out of fashion. 20 years ago I saw several large A-grade rigs powered with racks of Yamaha amps, which would be extremely unusual these days. Hill were also common around that time - now they have disappeared. A bit later, Crest were the "in" make. Now it's QSC and Lab Gruppen. Crown Macrotechs have always been there, but lost market share as people went for lightweights.
In the real world, engineering design always involves a degree of compromise. This applies to every commercial amp. Size, weight, cost, and reliability all bring about trade-offs.
This also applies to purchasing decisions. You may feel that MA5Ks sound best on your subs, but if fP6400s give you almost the same quality yet weigh 10kg as opposed to 25kg (each), you have to consider that as well, along with size, current draw, supply voltage tolerance, etc, especially if you need a dozen of them. Then again, you may feel that the lightweight sounds better than the heavyweight.
Have we ever had an amp shootout?
Post by: Tim McCulloch on November 23, 2006, 12:16:14 AM
| Dave Miller wrote on Wed, 22 November 2006 17:27 |
Have we ever had an amp shootout? |
http://srforums.prosoundweb.com/index.php/mv/tree/18200/1615 47/143/
Results here:
http://srforums.prosoundweb.com/index.php/m/5658/884/0///377 /?SQ=5152582855bdb422dc64a43eca061bb6#msg_5658
There's probably more....
Have fun, good luck.
Tim Mc
Post by: Ed Walters on November 23, 2006, 11:20:43 AM
While 1/3 power represents a worst case scenario and 1/8 power is max'ed out rock and roll, IMO an amplifier, if given a power rating by its manufacturer, should be able to deliver that rated power under a steady-state condition sufficiently long enough to measure with a basic test setup of load resistors and a voltmeter.
Whenever I test an amp, it's first stop, before I listen to it, is the bench, 1/3 power, for an hour. This tells me its thermal capacity, and if it is going to blow up if I rent it to a DJ, and if the amp is going to meet its published specs, which get checked next. Then I listen.
I am known for saying that the Lab Gruppen amps do not meet their published spec at 2 ohms. This doesn't mean they sound bad or don't work in live applications, it means LG should NOT publish an FTC-spec 2-ohm power rating, as the amps cannot deliver to that spec. This is about lies/marketing. It speaks to a corporate mentality that offends me to the point that it does not matter if the amp works, is reliable, sounds good, whatever, I still won't spec them. The rest of the people out there can say they're great/ bla bla bla, but the harsh reality is that LG lies, and this is fact, not opinion, easily verified by any tech.
So, to answer your question, live-sound amps never see 100% power in actual use. Does this exempt them from being capable of operating at 100% for a short time? No, not in my opinion.
Ed Walters
Post by: John Roberts {JR} on November 23, 2006, 12:27:01 PM
| Ed Walters wrote on Thu, 23 November 2006 10:20 |
Being the guilty party who originally posted about audible differences, and having a reputation re. test benches.... While 1/3 power represents a worst case scenario and 1/8 power is max'ed out rock and roll, IMO an amplifier, if given a power rating by its manufacturer, should be able to deliver that rated power under a steady-state condition sufficiently long enough to measure with a basic test setup of load resistors and a voltmeter. Whenever I test an amp, it's first stop, before I listen to it, is the bench, 1/3 power, for an hour. This tells me its thermal capacity, and if it is going to blow up if I rent it to a DJ, and if the amp is going to meet its published specs, which get checked next. Then I listen. I am known for saying that the Lab Gruppen amps do not meet their published spec at 2 ohms. This doesn't mean they sound bad or don't work in live applications, it means LG should NOT publish an FTC-spec 2-ohm power rating, as the amps cannot deliver to that spec. This is about lies/marketing. It speaks to a corporate mentality that offends me to the point that it does not matter if the amp works, is reliable, sounds good, whatever, I still won't spec them. The rest of the people out there can say they're great/ bla bla bla, but the harsh reality is that LG lies, and this is fact, not opinion, easily verified by any tech. So, to answer your question, live-sound amps never see 100% power in actual use. Does this exempt them from being capable of operating at 100% for a short time? No, not in my opinion. Ed Walters |
I will take your word for it that the LG doesn't meet spec. I have never had one of them on a bench. I do have a certain amount of experience with customers trying to bench test with inadequate rigs so I will also assume you have adequate mains service to your test bench, non-inductive load resistors, etc. I have seen major mains power rewiring projects required just to get design engineer's test benches adequate to design much smaller amps that those. You can't just throw a common variac on a sagging line and call that rated mains voltage. Unless you have one of those huge/heavy monsters they won't handle the current peaks, and I don't know if they even make something strong enough for the modern monster amps. I appreciate that one can argue that amps should make full rated power in the real world of shoddy distro's and once again we lack benchmark mains source impedance characterizations. Some amps (like PFC front end with class D back end are very mains friendly and I suspect that makes a difference in the real world.
For the record 1/3rd power is only worst case for simple class A/B amps which are never used above a couple thousand watts, because they are too big, heavy, and expensive. Once you go to multiple rail designs like class G-H they could game the voltage level of an intermediate rail to be quite happy at 1/3rd power. So worst case for an amp, depends upon the amp. In my experience most multi-rail amps are tweaked based on what works best in the majority of customer applications.
I am not aware of any reliable standard duty cycle for different musical genres or applications. Amp mfrs wish there were so they could dial in for that. If you have a series of benchmarks that work for you, keep doing what you're doing.
What I can see of the LG design approach looks pretty sound. It is not always trivial to translate an effective 230v design to 120v with the near doubling of mains current but the effort is worth the trouble due to the size of the 120V market.
In my experience most amp designers are conservative about power ratings and have nightmares about not being able to back up what they publish. At the power levels LG are dealing with I don't know if they are motivated to exaggerate.
Perhaps we can get a LG to road test and let you be the bench tester. You could document your findings and work with them to clarify any discrepancy. I am not disputing you claims but would sure like to give LGs a chance to review your tests and explain any differences in how they test.
happy turkey
JR
Post by: Ted Olausson on November 23, 2006, 03:38:50 PM
http://www.labgruppen.se/media/Testing_Procedure_v1.pdf
Post by: John Roberts {JR} on November 23, 2006, 05:25:12 PM
| Ted Olausson wrote on Thu, 23 November 2006 14:38 |
LAB has a paper that deals with amplifiertesting. http://www.labgruppen.se/media/Testing_Procedure_v1.pdf |
Thanks for the link... It reads like they have had issues from customers before about confirming performance with bench tests, and wrote this in response. It doesn't even address the things I was worried about for testing big dog amps (mains capability, etc).
I can see how consumers might not appreciate an amp that won't stay hot long enough for a distortion analyzer to settle and make a reading. "trust us it's good"...
I actually agree with them regarding most of the design philosophy.There used to be an IHF(?) test for dynamic headroom that was based on tone burst testing. I designed a small studio amp years ago that was a monster on tone bursts at 120W momentarily but only 35W continuous (actually 60W for 15 secs). It was damn loud for a 35W amp. So yes indeed music is not continuous tones.
I'm reminded of one of my old musings about a way to test gear based on the high repeatability of digital signals. The gravitation of test equipment to sine wave stimulus was because it was fairly easy to interpret the results. Any nonlinearities generated would be other than the pure tone stimulus. This was in my mind also a weakness of simple THD analyzers as they ignore phase shift and amplitude errors (i.e. frequency response), requiring a whole shopping list of specifications to adequately describe the accuracy of a signal path.
OK, back to my hypothetical test. Suppose instead of a sine wave, we blast the DUT (device under test) with some arbitrary prerecorded sample of music or whatever we want to use as a test signal. Since this is stored in a robust digital format we know exactly what we are looking for at the output.
While it might take some processing power to recalibrate for level but this is nothing we couldn't handle with modern DSP. So first we run a crude test that says the output is 99.XX % the same as the input signal... then maybe we could analyze the output further and determine we are deviating x % because of frequency response roll-offs at the extremes. Next the output deviates this other amount because of phase shift (maybe even qualify further between minimum phase and not.
With such an apparatus you test you amp with a sample of your music, and get a result that says it's 99.99% pure,,, or not. One number that tells it like it really is.
JR
PS: I'd buy one of these for $100
Post by: Mikael Holm on November 23, 2006, 08:47:23 PM
| Andy Peters wrote on Wed, 22 November 2006 07:47 |
Dunno where JJJ is right now, so I'll "help" |
Damnit! I better find a nicer picture for you guys
Post by: Mikael Holm on November 23, 2006, 09:02:41 PM
| Tim Duffin wrote on Wed, 22 November 2006 06:00 |
If a sine wave generator is not available, just play some jungle or bass music to test an amps LF performance--not Van Halen. |
I have done that but not close to clipping because (so far) i haven't found subwoofers that could take all the power my MC2 E45 amplifiers are able to produce. Maybe i just need to do some serial-parallel magic. Actually wouldn't be a bad idea at all because then those subwoofers might stay better in their linear operational range.
Post by: Nathan Lehouillier on November 23, 2006, 09:25:17 PM
So you have us all in suspense. You have a Venice and you
are not a fan of cheep QSC's. Please let me know what you
use for your best shows. If you tell us B&W's with Krell
amps and MIT wire. I will only be amused and truly think
you belong with the other morons I am running out of
business. Please feel free to put all your money into
what ever you want. I will bid against you win and the customer's will never know that your Superior amps weren't there. I think I am going out to get a new QSC tattoo.
Nathan Lehouillier
Owner
Kick Drum Sound & Lighting
1870 Rosewood St.
Green Bay Wi 54303
C- 920-471-9072
Post by: Mikael Holm on November 23, 2006, 09:35:51 PM
| Dave Miller wrote on Thu, 23 November 2006 01:27 |
I'd be very interested to hear if anyone has ever ran an amp at more than 1/3 power other than on a test bench. |
Chemical Brothers Bass Test ~ 1/2 duty cycle. I know people who have driven their amplifier at clip level (2ohm nominal load) listening that song. I also know people who have bottomed out cones and/or burned VC w/ that tune.
That kind of torture drives speakers into power compression very fast which means DJ tries to compensate but has nowhere to go. One just has to pray it's someone else's equipment under such abuse.
Post by: Mikael Holm on November 23, 2006, 09:45:15 PM
| John Roberts {JR} wrote on Wed, 22 November 2006 18:30 |
The slew rate wars ran their course several decades ago and (was it Matti Ottola) the champion of those early unconventional tests to parse out slew induced distortions |
Matti Otala, former head of design at Harman/Kardon (at age of 15 he built the first television set in Finland, in year 1954)
Post by: John Roberts {JR} on November 24, 2006, 09:52:03 AM
| Mikael Holm wrote on Thu, 23 November 2006 20:45 |
Matti Otala, former head of design at Harman/Kardon (at age of 15 he built the first television set in Finland, in year 1954) |
Thanks for the history. I didn't think he was an engineer. He and several others were pushing these slew related amplifier tests insisting they were a new unknown form of distortion.
They were generally dismissed by experienced engineers at the time. One that comes to mind was a Brit, Peter Baxandall who worked on RADAR circuit design during WWII. I guess if you can cleanly pass RADAR signals, audio frequency is trivial.
I am inclined to agree that audio frequency design is pretty well understood and we don't really need new exotic tests to parse out subtle flaws that are hidden from the view of conventional tests. I still like my 99.99% pure (digital comparison of output to input using complex music test stimulus). This is arguably a simplification while quite difficult as it includes all error sources (frequency response, phase shift, and nonlinearity).
JR
Post by: Don Boomer on November 24, 2006, 10:33:34 AM
| Mikael Holm wrote on Thu, 23 November 2006 20:45 |
I still like my 99.99% pure (digital comparison of output to input using complex music test stimulus). . |
I like it too ... now who gets to pick the "complex music" and will anyone else go along with it?
Post by: mel taylor on November 24, 2006, 10:36:40 AM
| Nathan Lehouillier wrote on Fri, 24 November 2006 02:25 |
Great, So you have us all in suspense. You have a Venice and you are not a fan of cheep QSC's. Please let me know what you use for your best shows. If you tell us B&W's with Krell amps and MIT wire. I will only be amused and truly think you belong with the other morons I am running out of business. Please feel free to put all your money into what ever you want. I will bid against you win and the customer's will never know that your Superior amps weren't there. I think I am going out to get a new QSC tattoo. Nathan Lehouillier Owner Kick Drum Sound & Lighting 1870 Rosewood St. Green Bay Wi 54303 C- 920-471-9072 |
Funny Nathan,
Glad you are doing well in your market.
Trust me, you won't be bidding against me, and I won't be coming to GB. My customers are not common.
Careful, tattoos are permenant. I wouldn't want to label myself for anyone, even the best.
Post by: John Roberts {JR} on November 24, 2006, 11:52:17 AM
| Don Boomer wrote on Fri, 24 November 2006 09:33 |
I like it too ... now who gets to pick the "complex music" and will anyone else go along with it? |
The beauty of it, it that you can use standardized test sources, or roll your own favorite worst case (like maybe board tapes from hell?) to use as your personal reality check.
We are not likely to ever have universal agreement on a musical test, and that will be a moving target as popular culture changes what is current. It seems to me (surely easier to speculate than execute) that algorithms to deal with phase shift and response errors due to channel bandwidth limitations occurring at the frequency extremes could be reasonably standardized. I suspect using wide band noise stimulus would generate larger reported errors than real music which should be inside the equipments passband (hopefully).
I would visualize a single overall (Ivory soap) test number 99.9% pure, and then the ability to drill down to identify how much of the deviation is amplitude, phase, or linearity errors. Such a test would have been discouraging (but IMO still useful) back in the early days of audio when we'd be doing good to hit 90%. Then the simpler more specific tests were useful to help circuit designers find error sources... this new test approach could still be used to perform basic tests by using simple test signals.
I stand by my position that analog design is pretty mature and the world doesn't really need new kinds of test to parse out unknown errors, but this kind of rolls all the others into a single result that should be easier to interpret.
Only caveat I would offer up front is that it might understate performance as the effects of minimum phase bandlimiting at skirts are generally benign. This could lead designers to deliver more bandwidth than is really useful to look good on paper.
JR
Post by: Mikael Holm on November 26, 2006, 12:33:47 PM
| John Roberts {JR} wrote on Fri, 24 November 2006 16:52 |
Thanks for the history. I didn't think he was an engineer. He and several others were pushing these slew related amplifier tests insisting they were a new unknown form of distortion. |
He used to be electronics professor before joining H/K.
Post by: Bob Lee (QSC) on November 27, 2006, 02:56:55 PM
What don't you understand about what I said? I'll be happy to explain if necessary.