REW and ARC Genesis results differ a lot. Frustrating me. Help?

[Snoochers]

I have two DIY subs and an Anthem MRX 1140. Long story short I've been very frustrated trying to get my system sounding the way I want. I've used ARC Genesis using best practices for measurement and ARC thinks things are great and REW does not agree. Look at these charts. This is at the exact same position. I am using UMIK-1 in REW and even used the ARC mic to see. This is using quick measure in ARC. The major difference is REW is using LFE and not just each sub separately but still. ARC sees like 1db variation and REW sees like 7db. I've bee chasing this for years and finally approaching my limit. The Anthem folk just say to ignore REW and ARC is great and say it's just some program differences or whatever but obviously there's a lot going on here.

I can't even EQ in ARC so things look flat in REW because options are limited and ARC already thinks it's perfect. I'd need to get a miniDSP or something to get REW looking good but then I reckon ARC will think things look bad. FYI this is a simple sweep in REW because ARC does a sweep too. A longer white noise moving mic method shows similar findings.

What should I do? Trust ARC? Trust REW? Reach some compromise? Is there another option?

 

[staticV3]

Unless you use the same mic in the same position, ARC and REW are unlikely to agree w/ each other.

 

[Snoochers]

Unless you use the same mic in the same position, ARC and REW are unlikely to agree w/ each other.

I literally used the same mic in the same position. That's the blue line in REW. But even if I didn't use the same mic in the same position, I wouldn't expect two reliable microphones to be off by 7db or more in the same position. They're supposed to be neutral or at least properly calibrated.

And OK let's say yes that using two different mics is the problem. Then what? How do I proceed? Trust REW with UMIK or ARC with ARC?

 

[JasonC331]

How is there so much output on the ARC past say 300hz, are you using any phase alignment on each ARG run? I did this too for a little while and just kept playing around until I liked it. Graph chasing can make you crazy

 

[Snoochers]

How is there so much output on the ARC past say 300hz, are you using any phase alignment on each ARG run? I did this too for a little while and just kept playing around until I liked it. Graph chasing can make you crazy

The subs output fairly high frequencies. Far along the graph I think it's just normal ambient noise.

I'm annoyed less by the chart chasing and more the fact that the charts are so different. ARC chart is perfect and REW chart with REW EQ is perfect, but why do they differ so much? Unfortunately I don't know how to use REW EQ since I am on a HTPC and don't have an external DAC and I need to use Atmos etc.

 

[OCA]

You can extract ARC measurements and drop them in REW with the attached script.

(remove .txt from the end of the file extension)

 

[ZolaIII]

Why don't you use same mic without any calibration in both? Old ARC mic's are worse than UMIK 1 but in bass there shouldn't be a difference. Newer one's are on pair with it. Idea behind ARC is to use PC for measurements and keep it better than what on AVR can be done. It tends to present it's better than it is. So limited adjustments you can make do to a difference of the two using ARC mic with calibration file for it on it. Ignore echo tail bottom, try to do calibration when ambient noise is minimal and on loud (upper 70's level and for that use UMIK 1 with cal in rew). Where is first room refraction (influenced by length in relatively normal rectangular room) to first try to deal with it? You can check that by REW simulator. That might actually be real difference between the two when I look at the graph's. Anyway when you lower it it should be less bumpy in REW and still not silk and smooth as ARC is trying to present it. You do a Butterworth (Q 0.707) crossover to a natural in room decay order and off you go. With REW you pretty much can do almost anything and combine things (PEQ a hell of bass and crossovers and do some tricks onto convoler) also not limited to exact chenel mapping for even more crossover or all digital self tuned ones for a lot of benefits passive can't do as cutting response below port tuning for woofer's or at Fs of a driver for example. How much you can at least direct to ARC I don't know as I don't use it.

 

[Snoochers]

You can extract ARC measurements and drop them in REW with the attached script.

(remove .txt from the end of the file extension)

Super cool! So I imported the 5 measurements for each of my subs that ARC took and my red line is my post calibration measurement in REW with UMIK-1. I think this makes some sense now. Basically ARC thought 15-20 was too high so is bringing it down. Same with 50-130.

My issue is.... what do I do now? ARC thinks it is doing the right things, but REW isn't very happy. Even if I do the mic moving method across the 5 positions I used for ARC it isn't happy.

 

[OCA]

Super cool! So I imported the 5 measurements for each of my subs that ARC took and my red line is my post calibration measurement in REW with UMIK-1. I think this makes some sense now. Basically ARC thought 15-20 was too high so is bringing it down. Same with 50-130.

My issue is.... what do I do now? ARC thinks it is doing the right things, but REW isn't very happy. Even if I do the mic moving method across the 5 positions I used for ARC it isn't happy.
View attachment 403936

Can you use custom target curves in ARC? I wrote that script but I have never used ARC in my life

 

[Snoochers]

Can you use custom target curves in ARC? I wrote that script but I have never used ARC in my life

No nothing custom which is kind of my problem. The program has only a couple sliders. It could be so much more powerful.

 

[OCA]

In your first post, second graph, Umik measurement looks like LFE channel is being lowpass filtered at 120Hz which is quite common with D&M gear when you try to manually measure LFE with REW. Anthem might be doing the same thing and you should try to find out how to disable it. It's possible to do with D&M.

Neither Umik nor ARC's mic can have reliable readings below 20Hz, you need very expensive mics for below 20Hz. Many receivers will show wobbles in that area as well. I wouldn't mind that too much, there's next to zero content in most tracks at these frequencies.

 

[Snoochers]

In your first post, second graph, Umik measurement looks like LFE channel is being lowpass filtered at 120Hz which is quite common with D&M gear when you try to manually measure LFE with REW. Anthem might be doing the same thing and you should try to find out how to disable it. It's possible to do with D&M.

Neither Umik nor ARC's mic can have reliable readings below 20Hz, you need very expensive mics for below 20Hz. Many receivers will show wobbles in that area as well. I wouldn't mind that too much, there's next to zero content in most tracks at these frequencies.

Thanks. What is D&M? And yeah I've changed the high frequency extension and played around with it a lot and it didn't help the graph tons. I can keep tinkering.

I'm ok with 20hz but a 6db dip from 50-130 doesn't sit right with me.

 

[MichaelJ]

This is very consistent with my experience with ARC. Even after EQing my subs flat using REW and a UMIK mic, ARC consistently shows a rising frequency response under 100 Hz. This causes ARC to attempt to flatten it, resulting in a declining measurement when checked with REW. Subjectively, this sounds terrible IMO.

After numerous ARC sweeps and even after swapping my mic at a local shop, I’ve resorted to manually EQing my subs with REW and my MiniDSP Flex and then defeating ARC adjustments of my subs.

In my opinion, ARC measurements in the bass region are fundamentally flawed. I can’t trust ARC’s measurements of the bass in my room and I struggled to get a satisfactory result in my system. In the end, I’m happy. But that’s only after a lot of finagling and experimentation that would have been easier implemented if they simply provided manual PEQ.

 

[OCA]

What is D&M?

Denon/Marantz

 

[Snoochers]

This is very consistent with my experience with ARC. Even after EQing my subs flat using REW and a UMIK mic, ARC consistently shows a rising frequency response under 100 Hz. This causes ARC to attempt to flatten it, resulting in a declining measurement when checked with REW. Subjectively, this sounds terrible IMO.

After numerous ARC sweeps and even after swapping my mic at a local shop, I’ve resorted to manually EQing my subs with REW and my MiniDSP Flex and then defeating ARC adjustments of my subs.

In my opinion, ARC measurements in the bass region are fundamentally flawed. I can’t trust ARC’s measurements of the bass in my room and I struggled to get a satisfactory result in my system. In the end, I’m happy. But that’s only after a lot of finagling and experimentation that would have been easier implemented if they simply provided manual PEQ.

Damn that's brutal. Good to know I guess. I don't know what to do. On the one hand Anthem has a good reputation and uses a 5 point measurement. On the other hand REW is used universally and is pretty open and doesn't like my measurements.

 

[MichaelJ]

I’m surprised this isn’t discussed more. However, it does come up in other forums where users often report that their bass sounds anemic after running ARC. Many seem to be swayed by the strong support for Anthem by the faithful and end up learning to "appreciate" the lackluster bass response post-ARC.

But the facts are clear: measurements from REW and UMIK consistently don’t align with ARC’s results. Personally, I trust REW and UMIK more because their measurements align better with my subjective experience and are widely recognized as a standard. Plus, UMIKs come with calibration files, which adds to their credibility.

 

[Snoochers]

I’m surprised this isn’t discussed more. However, it does come up in other forums where users often report that their bass sounds anemic after running ARC. Many seem to be swayed by the strong support for Anthem by the faithful and end up learning to "appreciate" the lackluster bass response post-ARC.

But the facts are clear: measurements from REW and UMIK consistently don’t align with ARC’s results. Personally, I trust REW and UMIK more because their measurements align better with my subjective experience and are widely recognized as a standard. Plus, UMIKs come with calibration files, which adds to their credibility.

DUDE exactly. I feel like i'm going crazy because I've talked to Anthem and an expert in ARC and both of them are shitting on me for using REW and not trusting ARC basically. I'm so pissed because if I want to go the REW route I can't even EQ enough to make REW happy.

 

[MichaelJ]

I can’t help but feel that one day, an Anthem engineer will discover a bug in the mic or software that’s causing these issues, and then they’ll casually upgrade ARC to accurately measure the bass region. Suddenly, the Anthem fanbase will be celebrating this "secret sauce" upgrade, when in reality, it will just be a fix for an obvious issue that anyone using a mic and REW could identify.

I don’t want to come off as trolling Anthem—it’s not something I typically do or wish to make a habit of. But it’s clear that something is off here. You and I aren’t the only ones noticing these discrepancies and feeling like we’re going a bit crazy over it.

 

[Snoochers]

I can’t help but feel that one day, an Anthem engineer will discover a bug in the mic or software that’s causing these issues, and then they’ll casually upgrade ARC to accurately measure the bass region. Suddenly, the Anthem fanbase will be celebrating this "secret sauce" upgrade, when in reality, it will just be a fix for an obvious issue that anyone using a mic and REW could identify.

I don’t want to come off as trolling Anthem—it’s not something I typically do or wish to make a habit of. But it’s clear that something is off here. You and I aren’t the only ones noticing these discrepancies and feeling like we’re going a bit crazy over it.

I 100% think it's the mic. The UMIK-1 is heavy and comes with a calibration file. The ARC one is lightweight and doesn't. Maybe it is paired to the receiver so the calibration is built-in but I don't know.

 

[JasonC331]

1.10.5 is now an official release, as usual, they give a few things fixed and say improvements Have you ever worked with anyone on your ARC file? There are a few people that tweak other's files for them (I do) but always remember ARC is showing a suggested target and tries to get there, REW and UMIK are much more accurate. I have asked Anthem several times for a calibration file with no luck. Maybe someone here could show a way to make one?