KRK Classic 5 Review (Studio Monitor)

[sarumbear]

With all due respect back, there is no such thing. A measurement takes time especially an in-room one which requires me to create an all new setup. Then I have to capture the graph and document it. Then I have post it here and answer challenges or questions about it. One question would be, "did you try this other combination of switches?" "How did the distortion change?" Or a million other questions.

It used to be that I could get started with measuring a speaker with NFS in about 10 minutes. Now it takes me 45 minutes to do all the static, non-NFS measurements such as CSD, driver response, distortion at each level, etc. Now you are asking me to measure the effect of the switch which would compound a bunch of these. These requests become standard demand for all future reviews compounding the time it takes to do these tests.

So no, there is absolutely, positively no such thing as "not time consuming." They are all time consuming and aggravating to deal with.

You miss read what I said. I said try the level controls “when you are listening the speakers (not measuring)” and tell us your thoughts like you do after adding the equalisation. My suggestion was purely on your subjective evaluations not the objective measurements.

Some of us value your subjective views too...

 

[sarumbear]

What's the point of doing so many measurements so quickly if they aren't representative of the real performance? I feel like this is a pretty good example of the danger of blindly following the measurements and making conclusions without critical thinking.

But they do represent the real performance. The performance of the speaker set as manufacturer labelled as flat is measured and the results are given. If you think following data is not correct nor it is the basis of critical thinking then I am afraid you will find this forum pretty frustrating.

 

[samwell7]

Thanks for doing this review @amirm!

I first cut my teeth on a set of Rokit 5 G3s and they didn't sound (as far as my memory serves) how the Classic measured, I distinctly remember a boosted mid-bass on the G3s which doesn't seem to be an issue here.

These measure better than I thought they would, although the HF is elevated and there are some discrepancies across the frequency response it overall looks fairly good with some toe in adjustment, EQ or playing with dip switches to get the sound 'right' for the end user.

I've added these to my 'to try' list, along with the DSP version which is available for not a great deal more money in Aus.

Can anybody tell me why the preference score is for far-field?
Not doubting the formula but the measurements seem better than a 2.8 preference score

 

[amirm]

You miss read what I said. I said try the level controls “when you are listening the speakers (not measuring)” and tell us your thoughts like you do after adding the equalisation.

You know what a pain it is to keep crawling in the back of a speaker to turn a switch on and off? It is a 100 times easier and more accurate to do that in software equalization. There is no reason to mess with these switches.

 

[amirm]

Can anybody tell me why the preference score is for far-field?

That is where the research came from. It uses a statistical model of how strong reflections are in typical listening rooms which naturally does not fit the use in near-field. There, a desk reflection would be far stronger for example than a floor in near-field.

I think directionally it is still correct so best to still look at it.

 

[samwell7]

That is where the research came from. It uses a statistical model of how strong reflections are in typical listening rooms which naturally does not fit the use in near-field. There, a desk reflection would be far stronger for example than a floor in near-field.

I think directionally it is still correct so best to still look at it.

Thanks for that! I hadn't noticed the note about far-field on other measurements.

Thanks again for your time and effort with these reviews.

 

[sarumbear]

You know what a pain it is to keep crawling in the back of a speaker to turn a switch on and off? It is a 100 times easier and more accurate to do that in software equalization. There is no reason to mess with these switches.

Thank you for the answer. We can’t have everything in life

However, I do thank you for making those tests available for free.

 

[MZKM]

Of if you can't, ask the company to give them to you.

Some will show the effects in the manual. I checked and KRK did not, even the descriptions for each adjustment is very non-detailed.

 

[MZKM]

I hadn't noticed the note about far-field on other measurements.

I’ll try to remember to do it in the future for monitors, I know I added the caveat when the LRS was measured.

 

[samwell7]

I’ll try to remember to do it in the future for monitors, I know I added the caveat when the LRS was measured.

Thanks for that, I thought it would've been the case of me just not remembering if I'd seen it before.

Thanks for your effort with the preference scores too, it's a great summary.

 

[infinitesymphony]

It might be. We have the KRK Rokit 8s, and have had to replace an amp module that started making random noises. And now the replacement is developing the same problem. The internet says we're not alone.

Amp board reliability has been a common problem with budget active monitors, many of which have components that are operating near the edge of rated specs in daily use. Sometimes if you can narrow down the problem components in time, you can substitute higher quality parts that last longer. Repaired a pair of M-Audio BX5As that way with a set of new power capacitors.

Here's a general guide for KRK Rokits:

https://www.gearslutz.com/board/gee...ixing-crackling-static-krk-rokit-peakers.html

 

[AnalogSteph]

Note an approx. +2 dB peak around 80 Hz despite the relatively cool garage - normal room temperature may easily add another dB. I'm guessing these are using the old trick of slightly boosting midbass to make up for things going down quickly below that (though -10 dB at 48-49ish Hz is by no means bad for a 5"). Place them on a desk at a wall, and bass would no doubt be abundant.

While treble could arguably do with turning down by a dB, it seems that levels around crossover actually are pretty much correct and most of the elevated response up there may be a function of the tweeter itself, which is odd.

All in all, the reputation of these little KRKs for a rather "hi-fi" sound seems deserved.

 

[YSC]

With all due respect back, there is no such thing. A measurement takes time especially an in-room one which requires me to create an all new setup. Then I have to capture the graph and document it. Then I have post it here and answer challenges or questions about it. One question would be, "did you try this other combination of switches?" "How did the distortion change?" Or a million other questions.

It used to be that I could get started with measuring a speaker with NFS in about 10 minutes. Now it takes me 45 minutes to do all the static, non-NFS measurements such as CSD, driver response, distortion at each level, etc. Now you are asking me to measure the effect of the switch which would compound a bunch of these. These requests become standard demand for all future reviews compounding the time it takes to do these tests.

So no, there is absolutely, positively no such thing as "not time consuming." They are all time consuming and aggravating to deal with.

Forget it then, the original request was asked solely due to the thing on T5V review showing what the -2db switch didn’t really tackle the problem in then distortion/in room graph and I thought that wouldn’t cost a lot of trouble. When it cost you so much time it’s obviously not worth to try it.
From what in stock setting it looks like they are up to almost the Adam’s category in 86db volume range. Should be nice enough in really near field entry level mixing or hifi stuff for what it cost

 

[JoeKickass]

...If you can't EQ and can't be bothered to learn, then you are not my target audience. You are living in a world of flawed sound reproduction given the mass impact of the room on the sound of the speaker.

You disagree? Get one for $140 and measure it yourself...

Well I use a PW Link for speaker correction, so any EQ is actually counterproductive. It's very easy and also has a quick measure feature to see the live frequency response (for example to see the effect of moving speakers or flipping trim switches)

 

[ROOSKIE]

Well I use a PW Link for speaker correction, so any EQ is actually counterproductive. It's very easy and also has a quick measure feature to see the live frequency response (for example to see the effect of moving speakers or flipping trim switches)

Huh? You are using an EQ, there are many names for it but that set-up is using EQ.

Basically you can see the directionality and the distortion and the frequency responses. That is pretty much all you need to decide if the speaker can be adjusted to taste. Plus Amir tests in his room for volume potential and other things that may come out in the subjective part.

If a system is close and your on the fence then buy and measure. You can always return the speaker if the PEQ/DSP/Room correction or whatever is unable to get you good sounds. In any case that dip switch can't do a better job than what a DSP can do.

 

[amirm]

Well I use a PW Link for speaker correction, so any EQ is actually counterproductive. It's very easy and also has a quick measure feature to see the live frequency response (for example to see the effect of moving speakers or flipping trim switches)

Your hope should be that the auto-eq you are using is able to sift the issues the speaker has among those overloaded by the room. In this case, if you set the target curve correctly, the fact that there is a tilt up response does not matter.

 

[paulraphael]

Amp board reliability has been a common problem with budget active monitors, many of which have components that are operating near the edge of rated specs in daily use. Sometimes if you can narrow down the problem components in time, you can substitute higher quality parts that last longer. Repaired a pair of M-Audio BX5As that way with a set of new power capacitors.

Here's a general guide for KRK Rokits:

https://www.gearslutz.com/board/gee...ixing-crackling-static-krk-rokit-peakers.html

Thanks so much for the link. I did see the black goop in there when changing the amp board, but did not realize it could be eating the speaker from the inside out. What a nightmare in there. This would rule out my ever considering this brand in the future.

 

[richard12511]

But this is a crucial measurement for this speaker that will define the purchase decisions of many... Why didn't we bother to optimize the settings before spending all day running the full measurement?

It should really be standard testing protocol to take a single fast on-axis measurement with different filter settings, and then use the best settings for the final extensive time-consuming 360/off-axis measurements.
.

Don't agree with this at all. For some speakers it might be trivial to find the best setting, but doing it for one speaker sets the standard and requires that he do it for all speakers(to be fair). My Genelec speakers have like 4 different bass tilts, 4 different desk bounce things, 4 different treble tilts. That's like 64 different combinations he'd have to try to then try and subjectively pick the "best" combination of settings. IMO, speakers should be measured in their default configuration. Making Amir try to subjectively select the best combination of settings to then measure creates way more problems than it solves.

 

[YSC]

Thanks so much for the link. I did see the black goop in there when changing the amp board, but did not realize it could be eating the speaker from the inside out. What a nightmare in there. This would rule out my ever considering this brand in the future.

this really looks like the downside for class AB amping, when I was deciding between Genelec and Focal I was afraid that focal shapes will have similar issue due to lack of big heatsink and being class AB

 

[Maiky76]

Don't agree with this at all. For some speakers it might be trivial to find the best setting, but doing it for one speaker sets the standard and requires that he do it for all speakers(to be fair). My Genelec speakers have like 4 different bass tilts, 4 different desk bounce things, 4 different treble tilts. That's like 64 different combinations he'd have to try to then try and subjectively pick the "best" combination of settings. IMO, speakers should be measured in their default configuration. Making Amir try to subjectively select the best combination of settings to then measure creates way more problems than it solves.

Hi,
HNY!

@amirm
How do you run your THD measurements?
Do select an axis run the FR/THD at two different levels? Is that it?
If so, it might be the right time to measure the built-in EQ, you could rerun the FR with the different settings.
I am not even sure if the desk EQ is needed...
From what I have seen with some speakers (check here: JBL 308p) it could get most of the job done with 0 effort.
When the effect of each setting is measured it is trivial to recalculate the scores etc.

Here are some thoughts about the EQ.

The raw data with corrected ER and PIR:
Score no EQ: 2.76
With Sub: 5.22

Spinorama with no EQ:

• Too much HF
• so bad for the port interferences
• Some other stuff also going on 230Hz and 290Hz

Directivity:
- Reasonable LW
- Better stay at tweeter height
- Better horizontal control than expected up to 5000Hz given the really minimal round-over around the Tweeter.
There is really no more excuses for not having a minimal waveguide...

EQ design:
I have generated two EQs. The APO config files are attached.

• The first one, labelled, LW is targeted at making the LW flat
• The second, labelled Score, starts with the first one and adds the score as an optimization variable.
• The EQs are designed in the context of regular stereo use i.e. domestic environment, no warranty is provided for a near field use in a studio environment although the LW might be better suited for this purpose.
• Once the LW is EQed to flat the Score is basically behaving as a 1.5dB trim of the HF
• Nothing can be blindly done with the interferences I am afraid.

Score EQ LW: 4.20
with sub: 6.62

Score EQ Score: 4.83
with sub: 7.19

KRK Classic 5 APO EQ LW 96000Hz
January042021-145927

Preamp: -0 dB

Filter 1: ON PK Fc 585 Hz Gain -0.67 dB Q 5.4
Filter 2: ON PK Fc 4277 Hz Gain -0.5 dB Q 5.49
Filter 3: ON PK Fc 5057 Hz Gain -2.25 dB Q 2.82
Filter 4: ON PK Fc 7360 Hz Gain -2.06 dB Q 1.66
Filter 5: ON PK Fc 17029 Hz Gain -2.94 dB Q 3.43

KRK Classic 5 APO EQ Score 96000Hz
January042021-145032

Preamp: -0 dB

Filter 1: ON PK Fc 584 Hz Gain -1 dB Q 7.93
Filter 2: ON PK Fc 2969 Hz Gain -0.74 dB Q 8.18
Filter 3: ON PK Fc 4810 Hz Gain -3.34 dB Q 2.07
Filter 4: ON PK Fc 7466 Hz Gain -3.06 dB Q 1.4
Filter 5: ON PK Fc 15783 Hz Gain -3.37 dB Q 2.44

On can add a High-Pass filter 50Hz, Q=0.7 that flattens the LF and decrease the output by 3dB @50Hz and 10dB at 30Hz.
The score drops a tiny bit (4.69 with Sub: unchanged at 7.2) but that's is worth a try.

Spinorama EQ LW

Spinorama EQ Score

EQ Score + HP

Zoom PIR-LW-ON

Regression - Tonal
Almost flat with the EQ score

Radar no EQ vs EQ score
Quite some improvements

The rest of the plots is attached.