This is great. This is what I was trying to do by Excel, manually. There was a lot of annoying things like different sampling rates, offsets, etc.

Definitely. For example my .aset put everything into centigrade and that might throw some one off if they track in F.

NICE!

(there may be a mixup, I always charge MUCH hotter than anyone so the two highest starters would me mine; MS1, MS2...it could also be one of the two is not selected to show)

Can you share the whle thing on the G share?

I recommend you down load the .alog you wish to compare from the drive and explore the artisan features. I'm not sure how the local .aset might distort the profiles being compared. And I'd course I am quite capable of mistakes.

Comparator doesn't generate something that can be shared - so what I posted above is just a screen shot.

aha, OK I thought someone could share this comparison as a whole. I'll try make one!



weird thing is that I added initial before the field that appears to be put in the legend, yet in the screenshot made by Artisan for HB those are gone (aha it's not the legend, it's unclear to me what is used to populate the field in the comparator and legend)

see, this is a screenshot

This is fascinating information - I would like to digest it further.

A question it brings to mind are the things most affecting the thermodynamics of the system:
My roasts are done with 300g charges and measured with TCs. I'd be very interested to hear what others are using, and if Huky users are working with perforated drums.

The two roasts that I shared were also roasted at a very low environmental temp (for me) of 0C/32F. This was also much lower than my practice roasts, so I wasn't quite able to get what I wanted out.

AFAIK I'm the only freak with a perforated drum, fitted with two RTDs, 400g batches

I learned it is very helpful to taste through various roasts of the same coffee and iterate the roast to try to reproduce a preferred cup.

I confirmed that I have a good ways to go to be more intentional in tasting and precise in roast iteration.

And I confirmed that my skills at cupping to predict whether a cup or an espresso from X coffee can be a winner or not are not that great. (My four favorite cups on my 9 sample cupping table did not include the coffee that made the pourover that pleased me the most.

I found our chosen coffee to have a winey ferment note to a strong degree that was difficult for me to appreciate. This exercise showed me that some other roasters using the same coffee could produce a coffee with less of the ferment and more fruit or chocolate, pushing the malbec character to the background.

My favorites on the cupping table were in order But my favorite cup was a pourover There are many reasons I would do such an exercise again, one being to have a chance to enjoy a cup like that. And to try to improve the correlation - on my cupping table the sample didn't shine into my top 4.

If a few of us or other home roasters are game I would propose another round in a few weeks in which I roast/send Burundi Long Miles rubirizi washed to a few profiles and we meet in a google meetup to discuss or cup together those samples.

(Sitting down to put together a summary I see that I don't have time tonight to decipher /transcribe precisely my cupping notes from months ago. But if anyone is interested in the full list, I could do so)

Quick summary - the coffee was all excellent, and I'd be happy to pay for any of it, other than my own...
Tasting so many roasts of the same coffee was interesting and challenging - i tried to cup 8 coffees in one session, and was totally lost - everything tasted the same. Eventually, i settled on groups of 4, with an additional control of my own roast.

Most interesting for me was seeing the different approaches. Since I've been on an espresso heavy coffee diet over the last few months, most of what I've been roasting had been for espresso. As a result, I've been aiming for slower and darker roasts than if i were in a filter first mood.

My best test roasts were quite similar to Chuck and Tim's coffees, giving deep body with chocolate strawberry and a crisp malic acidity. Superb espresso, either straight or in milk. No real ferment in either. The coffees that I sent were a little too fast in the first part of the roast, leading to a bit more aggressive acidity. Not what i was aiming for, but not too bad, and probably more interesting as filter.

Andrew and Flint's coffees both made excellent brews, but were too light for me to get a decent shot from. Hard to say if i preferred one, but i thought that Andrew's was a little sweeter, and Flint's a little brighter.

The coffees from Rick and Marcel were both kind of in the middle - neither filter not espresso roasts, but could be used for either. Not sure if omni roast is a perjorative, but i thought they were both excellent. Fruitier rather than chocolate forward as espresso, but not too aggressive in acidity.

Here's some data!

A few points:

1) Everyone measures events diffirently. I took the liberty to normalize them all to the way I do it (the right way, of course ). So dry end and first crack are set to 320 and 395 degrees BT.
2) Time = 0 is charge.
3) Sorry this is in Farenheit. I started converting, but am really tired of looking at Excel sheets.
4) RPB1 had a probe issue close to 1stC. For clarity, I deleted the bad data and interpolated with a constant slope in the dBT. The BT was left alone and you can see the small dip.
5) DM1 is obviously a bit funky. I think he forgot to hit "charge" until around turning point. Everything seems offset by 1.5 minutes or so.
6) Roast level was measured by Chuck on Roast Vision and given as Agtron equivalents.







The real value comes when viewing in Excel. You can go to the filter menu and hover over a single profile and see it highlighted against the others. I don't know if there is anyway to post that type of functionality here. I'm happy to share the excel files, though.