External Gear Ping not functioning


Here is what you claim:

And here is the reality Cubase users have been dealing with.
A very well known bug exists within Cubase. External effects negative ping cannot be calculated. Many versions have passed, many complaints have been made and no answer has been given.
I have been trying to work around this for a long time.
I would like an answer from someone from steinberg’s technical department as it impacts my workflow every day. Enough is enough. I would like to know if you intend to fix this or should I consider some other DAW?

Thanks in advance


How can you have a negative ping?

If you have software plugins enabled in your project, cubase will start delaying everything by many ms. The trip from digital to analog and back again is often many times faster than the whole project delay. For example lets say that you have so many plugins enabled that your project is delayed by 80ms. If your external gear roundtrip takes 10ms, then you have to tell the external gear that it needs to be -70ms. The numbers I used are huge just for the example. But you get the point.
That is something cubase has been strugling with for many versions now.

I would still appreciate an answer from any steinberg team member.

I think you got it wrong. The ping is just the roundtrip through your external gear and that’s it, Cubase will calculate the rest.

This deserves and answer honestly


No mate. I don’t have it wrong. Search external gear negative delay or simply check the link of the person who posted right after you.
There are people who have this problem and people who don’t. When you see it for the first time you simply can’t understand how this is possible. Exactly as you say. But if you press the damn ping it says 0 and when you print your effect and zoom in, you can clearly see that it is several samples earlier than the original. I don’t really care how or why. All I really care is that I can;t work with my gear.

So if for example you have a delay or a reverb you don’t really care, but if you are trying to parallel compress, then you have phasing issues.

It’s not likely a Cubase problem since all Cubase can do is report the round trip through the audio driver. Delay can only go in one direction and that’s either 0 (no delay) or some time in the past. Negative delay implies that something returns before it even happens - not possible. The confusion seems to arise for the fact Cubase automatically aligns all signals, this can either be the system delay (this channel is the hog) or it will be the system delay - the channel delay. Just because a channel with a 10ms latency is delayed by 70ms to match a total system latency of 80ms does not make it 10ms negative in relation to the system.

AFAIK, you should be measuring just the ping looping out and back through the interface channel, not through the gear, since that would distort the ping.

Also note: If you have a rather sophisticated interface with it’s own built-in mixer and effects - make sure you have that stuff properly defeated and out of the loop when you take the ping measurement or ALL bets are off. It would be way too easy to screw up your measurement with a internal mix loop.

Direct monitoring can produce a 0ms latency.
I have spent some time on this when I had the problem.
In my case totalmix (RME) was configured to direct monitor some channels, loop backs on others didn’t make it easier to figure out. In totalmix it is easy to set up multiple configuration, so I simply have one named Cubase that pretty much has everything set to default.
The whole dang thing should really be revised and made more flexible, but that’s another thread

The External FX ping delay is meant for DIGITAL outboard equipment that has its own AD/DA processing delay.

All other Delay Compensation is automatic, but this may depend on particular ASIO drivers and protocols for your particular brand of AD/DA.

No that is wrong, it does not matter if analog or digital, there will always be latency.
The ping goes to the output and is measured at the input, any analog or digital processing will take time. Cubase has no way of knowing that latency without it being measured using the ping.

No you’re wrong, I’ve tested this extensively.

Cubase does know the input and output latency. You can literally go look at it in your studio device setup window. Cubase knows the ASIO input latency, and output, and it knows the memory usage latency, and plugin latency.

What it does not know, as I stated, is the latency of a digital FX unit which is why External FX ping are set per EXtFX insert.

At least for me, this is all automatically calculated regardless of turning on or off ASIO guard, memory settings or plugin use. All of my analog non-digital mix gear returns to my DAW sample accurate with a ping of 0.00. test it yourself.

The only time latency is going to become a problem is if you’re using the DAW to distribute real-time monitoring to musicians.

Makes you wonder how the HW console guys handled a digital effects unit as a parallel send … They’s have to have a digital delay on anything mixed with it. More likely they’d only use it for a final print through…

I have been using external fx for a long time, and what you are saying here does not even make any sense.
Cubase knows the latency the driver reports, plugins and what not has no relevance in what the ping returns.
A ping is like in a submarine, sends out a signal and listens for that signal to return.
I would expect a digital fx, say a external digital compressor connected via sp/dif to have even higher latency than a analog compressor using analog in/output. If the ping returns 0ms something is usually wrong.
Test it yourself :slight_smile:
I use 3 different analog compressors using externalFX and do a lot of parallel compression. If my ping was set to 0ms my drums would turn to mush.

Huh? That’s exactly what I’m saying. A digital hardware compressor would return late because it’s own AD/DA time is not counted automatically by Cubase, thus you send ping through that digital compressor so that Cubase considers it.

There is no latency in an analog compressor, thus no additional compensation is needed for what Cubase already knows is happening and compensating for.

If your project has one second of latency, your recordings aren’t going to be one second late because Cubase knows what is being listened to is late, and it knows the input latency.

So why would anything analog not be compensated properly doing a round of DA into an anlog compressor and then back AD?

The pinger is for what Cubase isn’t including in its latency calculations.

I’m %90 sure. I thought the same thing you did, and then I took a day and my Cubase journal to test this out because I wanted to know, and no matter what I did to try and get things out of sync doing a round of DA back to AD, it was always sample accurate. I maxed all latency inducing settings and plugins, and it always came back sync’d with the ExtFX delay set to 0ms.

First of all I am using an Apollo X8p and everything is set to bypass. Even the unison preamps. So that is no issue. Same thing used to happen with every other interface I used before that.
Also, regarding the gear I am pinging through. It makes no diff. Even if I ping through my patchbay and back, without any gear, the same thing happens.

Negative delay is a fact in other daws so please try to be open. It doesn’t mean that something returns before it even happens. Think about it. If everything I am saying is my problem and noone elses then why are there so many topics about Cubase not reporting ping correctly. Or even if it manages to report it, it doesn’t compensate properly.

This bug is ANCIENT! Its not a matter or debate if it exists. Its only a matter of WHEN THEY ARE GOING TO FIX IT!

Here!!! :




Have you tried setting it to 0.0ms delay, and sending out a click or test tone and recording it to see if the recording is aligned?

I did try looking up those gearslutz post concerning “negative” delay - some of those are indeed quite old. Those threads are claiming that certain tracks are being recorded earlier than they should be. They seem to be confined to specific audio interfaces. Somebody quotes RME as saying “negative delay values are possible, Steinberg is aware…”

There you go - RME driver reporting a “negative” delay and Cubase ignoring it. Both companies have an argument here.

For one, “negative” delay can only be a relative measurement, not an absolute measure since it is difference between two measurements. For a relative measurement you can have either the same, more, or less. Which one is the reference of the two measurements determines if this relative measurement is negative or positive.

Delay itself though, is a absolute sum, you either have some of it, or none at all. No reference is needed, since the absence of any delay is simply 0.

RME is reporting a negative delay, they say. Really though, compared to what? If compared to 0 that would be logically impossible, for obvious reasons. So Cubase appears to have adopted the stance that, since negative system delays are not possible, ignore any driver that reports such a thing.

I would be willing to bet a wooden nickel that the early times on these tracks are precisely that “negative” delay*-1 …