AR.com

[fi-guy]

Please advise.

I have a first-time project in mind. I have been looking at this DIY thing passionately since 2003. The Ed Frias Design for AR.com sounds promising.

If you've built these and like them please reply why, and, what you have compared them to. I have read reviews claiming better than anything up to $1000 in the commercial market.

PROBLEM: I have heard speakers in the 750-1200 range that I absolutely hate!!! Contrarily, I have head some in the same range that just shine!

Please advise.

[Zaph]

This topic comes up from time to time.

I'd advise against this kit. It is a poor crossover design and the tweeter is a poor performer. There are several updated crossovers. The update released by Ed Frias does not fix any of it's problems. There are however a few other crossovers that are improved, one of which is: http://www.gr-research.com/edspeaker.htm

Even with it's poor design, people will line up to tell you how great it is. Don't fall for that hype, you can do way better. If you want to use that woofer, Wayne J at http://www.speakerbuilder.net/ has a much better design.

[MCH]

Hmmm!
I guess I fell for the hype. I've built many complete sets of these, 5 speakers for surround, for friends and family. I still haven't heard anyone say how terrible they sound. They all seem to be pleased with their sound. My brother has Paradigm Studio 80s and prefers the ARdiys over them.
I personally love the sound of these speakers with a Tempest subwoofer. But then what do I know. Seriously, speaker sound is very subjective. Everyone has their own likes and dislikes.
When there are many positive reviews; I won't let a couple of negative feedbacks be a determining factor.
I'm glad I ran across the ARdiys.

[Lusso5]

I built a set of these for a friend, and later one for myself (I had the 812687s on hand). I think there is a good price-performance ratio to these.

I listened recently to the Polk LSi7s and RTi6s. The LSi7s were an easily noticable improvement over the ARs, with a little less bass extention; they were marked $850 and I would consider them "better" than the ARs. The RTi6s were pretty dissapointing, and were marked $400; I would say they do not compare to the ARs.

I'm not impressed with the 812687 tweeter, but think highly of the 850122 woofer.

If I were to build another bookshelf, I would do something with the Extremis 6.8, when a good crossover gets published.

Other that that, you might consider:

Dennis' CAOW1
http://murphyblaster.com/content.php?f=CAOW1.html

One of Tony's bookshelf designs
http://home.hetnet.nl/~geenius/

[Zaph]

Originally posted by MCH
I won't let a couple of negative feedbacks be a determining factor. I'm glad I ran across the ARdiys.

How about actual measurements.

[Lusso5]

Originally posted by Zaph

How about actual measurements.

I, quite honestly, dont notice that valley, or peak in my ARs. That doesnt mean it isnt there, and I dont have measurements. I do wonder if they are exasperated some by the tweeter not being flush-mounted in Danny's tested speaker.

[MCH]

....."How about actual measurements"......

Again I say listening is very subjective to what sounds good. If you need measurements to tell you what they sound like, that is your problem. For my sound tastes and budget, I think they are an excellent build.
I think it is impossible to ask for what sounds good. Every individual will have a different preference. You can get into endless debates on wires, electronic breakins, yada yahda.
From my prespective and listening pleasure I'm extremely happy with my ARdiys.
the end!!!!

[Zaph]

Originally posted by MCH
....."How about actual measurements"......

Again I say listening is very subjective to what sounds good. If you need measurements to tell you what they sound like, that is your problem.

My ears work fine, thank you. I don't need measurements to tell me how a speaker sounds, I merely posted that response curve as information. I can understand how that would make you feel bad. I indirectly said that you lack the experience to identify a poor sounding speaker. I'm sorry. But the fact is that the AR.com diy kit is often the first kit built by the people recommending them, and all they have to compare with is shoddy commercial products. That's why I mentioned that people will line up to tell you how great this kit is.

If they sound good to you, I'm glad you're happy. I guess people will just have to decide for themselves.

If anyone wants more opinions on the ar.com speaker, Madisound might be a good place to ask. There are many people over there with a lot more experience and design knowledge than me. They will echo my sentiments.

Originally posted by Lusso5
I, quite honestly, dont notice that valley, or peak in my ARs. That doesnt mean it isnt there, and I dont have measurements. I do wonder if they are exasperated some by the tweeter not being flush-mounted in Danny's tested speaker.

The top octave peak is just a result of the poor tweeter. It adds excessive sizzle to the sound. The dip at 3khz is a little too low to be a tweeter surface mount diffraction issue although there is probably some effect. It's definitely related to poor phase tracking between the tweeter and the woofer.

[DcibeL]

Perhaps if the tweeter was changed in this particular design, we would see better results without having to make an over elaborate and likely expensive crossover? The Tweeter used in the PeeCreek looks to be a nice one, but if I remember correctly it is also quite a fair bit more expensive.

I had a look at the tweeter and it doesn't actually look that bad. I don't have experience with any high end gear besides my own, so all I can suggest is the tweeter I am using, the 811815 from Peerless, which is being replaced by the 811830, a low compression version of the same driver.

[HeatMiser]

It seems to me that kind of dip probably actually contributes to the favorable reaction these speakers get in some cases; a big suckout right there eliminates any chance of the really annoying nasality and "shout" you get in a lot of less-than-capable speakers that shoot for being revealing but fail. And it probably lends the impression of a richer midrange/upper bass compared to typical commercial designs that have no baffle step compensation and sound thin and weak as a result. This at a cost in vocal intelligibility and general presence in the upper mids.

My own current speakers are like this, which is why I hang out here trying to learn enough to build something better. They sound deceptively good on rock and pop and electric jazz, but they are almost useless for home theater (I have to strain to hear dialog) and seem seriously compromised for classical and acoustic and other types of music where you know what it is "supposed" to sound like, and it unfortunately doesn't sound like that.