which is the best hardware for peak limiter or brickwall limiter ?

[Lion]

Which HARDWARE do you use for reduce peak and increase the level ?

Thanks.

[carlsaff]

In the hardware realm, I'm pretty sure the Waves L2, long in the tooth though it may be, is still the most popular option.

I'd love to hear about hardware alternatives, too, though. I've considered hardware peak limiter options from time to time, but I keep coming back to software limiting, which gives greater flexibility and in my opinion can yield better tone than the L2. It's also nice to have the limiting for a master be something that is not "fixed" (i.e. not part of the digital capture from the analog chain). Then it's settings can be easily tweaked (even automated) after capturing audio from the analog processors. This saves time, as one client's "too much limiting" is another client's "not enough limiting."

[Lion]

Originally Posted by carlsaff

In the hardware realm, I'm pretty sure the Waves L2, long in the tooth though it may be, is still the most popular option.

I'd love to hear about hardware alternatives, too, though. I've considered hardware peak limiter options from time to time, but I keep coming back to software limiting, which gives greater flexibility and in my opinion can yield better tone than the L2. It's also nice to have the limiting for a master be something that is not "fixed" (i.e. not part of the digital capture from the analog chain). Then it's settings can be easily tweaked (even automated) after capturing audio from the analog processors. This saves time, as one client's "too much limiting" is another client's "not enough limiting."

I see what you want to say, for the moment i work with software limiter (L2, L3, L2007, TC Brickwall Limiter from the powercore, and each have a different sound, depend on the music i work) but i think that a hardware Limiter is better than the software ?!?

[carlsaff]

Maybe. But, as an example, it is a topic of much debate whether the hardware L2 sounds better than the software L2. Waves claims the two are identical. Others claim a clear difference.

If you're getting the bulk of your level other ways, then you won't be doing that much digital limiting, anyway. There are many software limiters that can do up to 4db of limiting very transparently.

[Lion]

Originally Posted by carlsaff

Maybe. But, as an example, it is a topic of much debate whether the hardware L2 sounds better than the software L2. Waves claims the two are identical. Others claim a clear difference.

If you're getting the bulk of your level other ways, then you won't be doing that much digital limiting, anyway. There are many software limiters that can do up to 4db of limiting very transparently.

It's a good point of view.
Thanks

[MASSIVE Master]

I have a modified Aphex Dominator II (722) which I love, but I tend not to use much.

The HEDD takes a pretty good beating also depending on the settings.

[Lion]

Originally Posted by MASSIVE Master

I have a modified Aphex Dominator II (722) which I love, but I tend not to use much.

The HEDD takes a pretty good beating also depending on the settings.

I was thinking that the HEDD didn't have any peak limiting system.
Does the "tape emultion" work like a peak limiter ?
What do you use for the end of the chain (to set to 0dB), a software brickwall limiter or just the HEDD?

[Branislav]

Originally Posted by carlsaff

In the hardware realm, I'm pretty sure the Waves L2, long in the tooth though it may be, is still the most popular option.

L2 is arguably a hardware limiter, as it performs limiting in software, though in an external box. If you were looking for an analog box, than Dominator is the best I tried, but software limiters are better for fast brickwall limiting and if you still want a box, than go for Weiss, nothing comes close to it.

Branislav

[Darius van H]

Originally Posted by Lion

I was thinking that the HEDD didn't have any peak limiting system.
Does the "tape emultion" work like a peak limiter ?
What do you use for the end of the chain (to set to 0dB), a software brickwall limiter or just the HEDD?

The HEDD processing can act as a limiter, but it's important to realize it's a top-down process. As you increase the processing, the mamimum level decreases, with the peaks being rounded downwards.

Many people make the mistake with the HEDD of thinking that they can push the input of the HEDD to "work" the limiting harder. But what's happening is that the A/D is being clipped, then the processing is rounding off those clips (purely in the digital domain) - essentially masking (intellectually and perhaps audibly) the A/D clipping distortion with even more distortion.

While clipping an A/D and adding distortion is a valid and widely used tool in mastering these days, it's important to be aware when and where it's being done!

What do i use for limiting? Clipping, L2, GClip, that's about it. But to be honest, loudness is generally achieved before the limiting stage thru EQ and compression and other tricks.

[Bob Yordan]

I am kind of fond of my Weiss DS-1, Manley Slam & Vintage
Design Ha2-16 (custom built OTB).

But I have nothing more to compare to except UAD Precision Limiter & Sonnox Limiter in the digital domain.

[Lion]

THANKS for the HEDD explications, and the others posts.

[jayfrigo]

My favorite hardware limiter at the moment would have to be the t.c. 6000 with the newer MD4 algorithms. However, it's not the only one I have and use, nor is it the only one worth using. It plays well with others, but if I had to pick only one, I think the t.c. 6000 would be it.

[Lion]

Originally Posted by jayfrigo

My favorite hardware limiter at the moment would have to be the t.c. 6000 with the newer MD4 algorithms. However, it's not the only one I have and use, nor is it the only one worth using. It plays well with others, but if I had to pick only one, I think the t.c. 6000 would be it.

Hi,
have you ever compare the hardware vs software version of the TC M6000 (MD3 & brickwall limiter) because i have a Powercore with MD3 & brickwall (it works great) but want to know if there is a big difference (for the sound not for the capability which is better on the hardware version)between hardware & software version of the M6000.
P.S. : It would be great if Tc release a MD4 for Powercore but i think it's impossibe due to the big DSP power that a MD4 need...

Thanks,
Lionel.

[jayfrigo]

Originally Posted by Lion

Hi,
have you ever compare the hardware vs software version of the TC M6000 (MD3 & brickwall limiter) because i have a Powercore with MD3 & brickwall (it works great) but want to know if there is a big difference (for the sound not for the capability which is better on the hardware version)between hardware & software version of the M6000.
P.S. : It would be great if Tc release a MD4 for Powercore but i think it's impossibe due to the big DSP power that a MD4 need...

Thanks,
Lionel.

There truly is a big difference between MD3 and MD4, both in sound quality and features and flexibility. I really didn't use the 6000 nearly as much as I do now before MD4 came out. I think MD4 could eventually be released as software as DSP power is getting faster and cheaper all the time. It's no more power hungry than the MDW (Massenburg) EQ. I think they'll wait a while primarily from a marketing perspective, but perhaps powercore is a little DSP shy, in which case they might come out with an updated version at some point with some more software. But we'll probably get MD5 (not the checksum) for the mainframe if MD4 goes native!

[Lion]

Originally Posted by jayfrigo

There truly is a big difference between MD3 and MD4, both in sound quality and features and flexibility. I really didn't use the 6000 nearly as much as I do now before MD4 came out. I think MD4 could eventually be released as software as DSP power is getting faster and cheaper all the time. It's no more power hungry than the MDW (Massenburg) EQ. I think they'll wait a while primarily from a marketing perspective, but perhaps powercore is a little DSP shy, in which case they might come out with an updated version at some point with some more software. But we'll probably get MD5 (not the checksum) for the mainframe if MD4 goes native!

I think too that the MD4 is better than the MD3 (because it come whith a lot of improvements) but i would like to know if there is a difference (for the sound) between the hardware & software version of the MD3 (if you have made a comparison). Maybe, i think it's the same because it's in digital domain, is it right ?

Thanks.

[jayfrigo]

Originally Posted by Lion

I think too that the MD4 is better than the MD3 (because it come whith a lot of improvements) but i would like to know if there is a difference (for the sound) between the hardware & software version of the MD3 (if you have made a comparison). Maybe, i think it's the same because it's in digital domain, is it right ?

Thanks.

I haven't heard them side by side, but I would expect them to be very close, if not identical. There's no reason that the same software running on different hardware can't be just as good. Then again, I'm assuming things, so this is not really a reliable answer.

[lucey]

to me it's all about small doses in combination. i clip the converter a touch, D->D with the Hedd and if necessary, L2 and or Soft Clip on the Alpha.

No one approach is best, or should be looked at for more than a touch.

[Lion]

Originally Posted by jayfrigo

I haven't heard them side by side, but I would expect them to be very close, if not identical. There's no reason that the same software running on different hardware can't be just as good. Then again, I'm assuming things, so this is not really a reliable answer.

Thanks.
And hope they will release a MD4 for powerore or Protools TDM system quickly, like you say, if it's not more power hungry than the MDW EQ (great digital EQ, i love it) it should be possible...

[Lion]

Originally Posted by lucey

to me it's all about small doses in combination. i clip the converter a touch, D->D with the Hedd and if necessary, L2.

I dont think any one approach is best, or should be looked at for more than a touch.

Thanks Brian. I agree with you that any one approach is best, and depend on the source / project / style / client...
I 've seen that a lot of people use clipping or soft sat/limit in the A/D converters... In some case, i personally use the soft limit of my Apogee Rosetta 800 (it's not the best but make a good job for the moment, i will probably move for a Lavry blue AD/DA or other the next month) and then after a digital brickwall limiter (TC BRICKWALL, L2,...)

[Jake57]

I'm wondering, given that Algorithmix has recently announced Dynamix, which features among other things a parallel mode, if TC would now consider releasing MD4 on Powercore.

[yareck]

Originally Posted by Jake57

I'm wondering, given that Algorithmix has recently announced Dynamix, which features among other things a parallel mode, if TC would now consider releasing MD4 on Powercore.

Did someone try this new Algorithmix ? Any thoughts?

[Andy Krehm]

Originally Posted by Lion

Thanks Brian. I agree with you that any one approach is best, and depend on the source / project / style / client...
I 've seen that a lot of people use clipping or soft sat/limit in the A/D converters... In some case, i personally use the soft limit of my Apogee Rosetta 800 (it's not the best but make a good job for the moment, i will probably move for a Lavry blue AD/DA or other the next month) and then after a digital brickwall limiter (TC BRICKWALL, L2,...)

A few years ago, I used the Apogee PSX-100 (w/ Big Ben) on many, many masters. I also used the Soft Limit most of the time and found it very useful.

When I decided to upgrade my converters, I still wanted to use a "soft limiter" but wanted it to sound better. I did some research and discovered that the Apogee Soft Limiter was an analog limiter.

I then discovered that none of the high end converters that I was interest in had an analog limiter. They were all digital.

I was just so used to the soft limit that I still really wanted to have one available after ditching the PSX-100 so I looked around and found that Pendulum was about to release an analog peak limiter, the PL2. Greg the designer told me it would be exactly what I was looking for so I put myself on the waiting list and got one of the early units.

I then bought a round trip of Lavry Gold converters. I set up a test so I could A/B the Apogee Soft Limit against the PL2 and it was no contest. The PL2 was much better.

This was a couple of years ago and I still use the PL2 just before the ADC most of the time and seldom use the Lavry Gold SoftSat, although it has been useful once in a while.

[lucey]

That's a nice chain Andy. I didn't need that PL-2s power/sound for my chain but it's a great unit.

[TanTan]

i use a DCS 904 as a peak limiter when pushing it hard, love it !
for software the L3 multimaximizer is working good for me for 0.5-1Db GR

[Ben F]

In my blind listening tests using 24bit word length the L2 hardware sounded better than the L2 plug-in. The plug-in sounded slightly 'dry' and lacking depth. Didn't do a null test before anyone asks.

The Weiss DS-1 is the current limiter of choice. I was actually impressed by the McDSP ML4000 limiter, and the UAD-1 precision limiter.

[bob katz]

Originally Posted by Ben F

In my blind listening tests using 24bit word length the L2 hardware sounded better than the L2 plug-in. The plug-in sounded slightly 'dry' and lacking depth. Didn't do a null test before anyone asks.

The Weiss DS-1 is the current limiter of choice. I was actually impressed by the McDSP ML4000 limiter, and the UAD-1 precision limiter.

I just want to point out that both Alan Silverman and I have done null tests and the L2 plugin pretty much nulls against the hardware. So I suspect some placebo effect going on. I've been assured by Waves that the algorithm is the same. HOWEVER: The hardware L2 has one significant advantage: UNLINKING. It is possible to run the L2 hardware unlinked, which is a meaningful sonic advantage to my ears. Where a listener might hear some clamping effect when linked, when unlinked, the channel that doesn't gain reduce masks some of the clamping (release time artifact) of the other channel.

In fact, there is NO current hardware made today by Waves that has the link switch. The Max BCL, which is the current model, does not have a link/unlink switch. Nor does the plugin.

[Ben F]

Originally Posted by bob katz

I just want to point out that both Alan Silverman and I have done null tests and the L2 plugin pretty much nulls against the hardware. So I suspect some placebo effect going on. I've been assured by Waves that the algorithm is the same. HOWEVER: The hardware L2 has one significant advantage: UNLINKING. It is possible to run the L2 hardware unlinked, which is a meaningful sonic advantage to my ears. Where a listener might hear some clamping effect when linked, when unlinked, the channel that doesn't gain reduce masks some of the clamping (release time artifact) of the other channel.

In fact, there is NO current hardware made today by Waves that has the link switch. The Max BCL, which is the current model, does not have a link/unlink switch. Nor does the plugin.

Bob you're not supposed to give scientific proof! Let me believe my hardware is better. I actually suspect that it's because of going AES via the LYNX and Prism convertors. So yes the test was flawed, but still a good enough reason to avoid the software.

[stellar]

wow I just skimmed this thread and every unit discussed or mentioned is a digital unit..........not what I was expecting!

[jayfrigo]

Originally Posted by bob katz

I just want to point out that both Alan Silverman and I have done null tests and the L2 plugin pretty much nulls against the hardware. So I suspect some placebo effect going on. I've been assured by Waves that the algorithm is the same. HOWEVER: The hardware L2 has one significant advantage: UNLINKING. It is possible to run the L2 hardware unlinked, which is a meaningful sonic advantage to my ears.

I'm beginning to suspect that the metering may have something to do with the impressions. It may be that .5 to 1.5 dB is what sounds best on L2, and that it is easy to read on hardware, but the metering in the software reacts differently. It may lead you to apply more than you think you are applying. Just a guess...

I always thought they should be pretty much the same, but recently used the TDM plug to demonstrate something to somebody, and was really surprised that it sounded much worse than I was expecting (and had remembered hearing on this song previously using hardware). As to placebo and expectation, I was expecting it to be fine, but it wasn't. Hmmm, don't know what to make of it. Maybe it's just that it's been a long time since I've used L2 as a primary and it's worse than I'd remembered? Either way, a question I had thought was closed I'm now open to investigating again.

[lucey]

Originally Posted by bob katz

I just want to point out that both Alan Silverman and I have done null tests and the L2 plugin pretty much nulls against the hardware. So I suspect some placebo effect going on. I've been assured by Waves that the algorithm is the same.

When you say "pretty much" what's the difference? Very low numbers are audible, as you must know.