The FrogPad
 
A forum to get help or talk about Roastmaster…or anything else coffee.

Show Posts

This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.


Messages - Danny Hall

Pages: 1 ... 21 22 [23] 24 25 26
331
Support / Re: Great app
« on: January 29, 2013, 01:27:48 PM »
Almost  :)

332
Support / Re: Great app
« on: January 27, 2013, 05:35:07 PM »
Hi Truk

Great to hear from you, and I really appreciate the kind words!

Glad you were able to find some help - the site is in a bit of disarray. Right now I'm switching things around in the background so that help and documentation files will be easy to add, maintain and most importantly - find. Also to create a place for the upcoming tutorial videos, which are long overdue. Hopefully the new site will be up and running this week.

Yes - curves are your means to store any variable that changes throughout a roast. They can be one of 2 types - "Reading" curves for temperature readings, and "Control" curves to represent control information, like Gas flow, fan, heat settings, etc. In the roasting console, to add a value to a curve, simply make sure it's selected (tap the graph to select through each curve in succession), then tap the digital readout near the gauge to create or edit node data. That will either A: create a new node at the current roast time if no node is selected, or B: alter the value of the selected node if a node is selected.

Curves belong to one of three things - according to where you create them: A roast, a profile or a program. Whenever you use a profile or program that contains curves, those curves display in the roasting console and analyzer alongside any curves you have created directly in that roast. The list view of the roasting console will tell you the location of the curve.

Yes - you are correct, if you delete a curve, it's gone for good. I've tried to instill warnings throughout the app where you might inadvertently delete a curve that is being used by multiple roasts (such as in a profile or program), so Roastmaster should give appropriate warnings. The way that this all works is rather complex, I've tried to make it as intuitive as possible, but still flexible. The goal is to have it "just work" without a lot of thought.

Yes - In a basic roast, you create the curves you want. Here's an example...

Using curves on a per-roast basis:
With my Quest roaster, I might create 2 control curves - 1 for heat, and 1 for fan. These would store what settings I use on the machine. I'd enter the value of the control immediately after starting the roast (the first value you enter via the roasting console is always time-stamped at 0 seconds) and then adjust the appropriate curve whenever I change a control on the machine. I'd also create 1 Reading curve - to store what temps I recorded throughout the roast, and input readings whenever I felt appropriate, usually 1 immediately at bean drop, then every 30 or 60 seconds as I felt appropriate. (These curves will later be very informative in the Roast Analyzer environment - since i can quickly swipe through past roasts graphically, and see cupping data and curves for each past roast in a flip-book type of UI.)

Now, that's well and good, but doesn't provide a "workflow" so to speak, and there's lots of manual entry with each roast. That's where profiles and programs start to help. Profiles and programs are identical in their function. The Behmor uses both, so I included both. You can pick whichever moniker you like and ignore the other.

Using curves with Profiles and Programs
So, to start off... If I wanted to eliminate the extra steps of creating curves for each roast, according to the Quest example above, I'd create a profile, then I'd define 3 "curve templates" - 2 control and 1 reading - in that profile. Then, whenever I use that profile in a roast, those 3 curves would be duplicated in that roast automatically as blank curves, and I can avoid those extra steps in each roast.

Now as nice as that automation is, you'd soon think "I haven't yet done anything to help me shape my future roasts based on my past roasts." So, as you roast, and find combinations you like and want to duplicate, you can copy and paste the corresponding curves from a particularly good roast you like into to a new profile or program. Then, whenever you choose that profile in a new roast, those curves will display in that roast as well. In this scenario (with the Quest example), the "Control" curves would be your guide as to what to control on the machine, and the "Reading" curve would be the curve you're trying to match, so you would make 1 "Curve Template" in that profile as well - of a Reading curve type. That way, in the roast you'd see the control curves you pasted to inform you of machine settings to achieve that roast, as well as the past reading curve you pasted with the temps you are trying to match, and finally, the new reading curve overlaid that will be automatically created in the roast for you to record temp readings to as the roast progresses.

You'd probably soon find you want to experiment with this "recipe" as you roast, and tweak machine settings to see what outcomes you get. Anytime you alter a curve that belongs to a profile or program, in the context of a roast, Roastmaster will create a snapshot in that roast, so that the original master profile remains untouched, and the new snapshot becomes editable - allowing you to tweak the curves of the profile as you roast. This happens automatically in the roasting console - whenever you try to edit a curve belonging to a profile or program, you'll hear the shutter-click camera sound and the profile name will display with an asterisk to denote that it's a snapshot. Otherwise, if you're using the roast analyzer full screen, you'll need to manually create the snapshot first, then edit it.

One final note:
You'll soon realize that Control curves are directly tied to the weight you're roasting, so cannot be interchanged among roasts of different weights. Most home roasters will be bound to a machine that only has a heat control, with no way to measure bean mass temp, so this scenario will be ideal for them.

If you have a roaster that allows to read the bean mass temp, you'll probably prefer to roast by bean temperature readings alone - adjusting the controls to keep your new reading curve in sync with your target reading curve. That's the great thing about temp reading curves - they are not weight-dependent. In other words, the same profile curve that works for 100g, would work for 100kg.

Hope this all helps! Let me know if you have more questions, and thanks for the great feedback!

Kind regards
Danny Hall

333
Roasting / Re: Roast Number 400!
« on: January 08, 2013, 12:52:35 AM »
Love it! You're roasting faster than me, so I'm thinking you'll pass me in the spring!  :)

334
Support / Re: Facebook error
« on: December 17, 2012, 01:14:53 AM »
Hi Micah

Try turning your Facebook and Twitter settings off, then back on again in the device Preferences.

I rewrote Facebook in this update to function solely on the iOS SDK, instead of the Facebook SDK. I ran into what may be the same problem with Facebook during development, but could never find an issue in the code. Every time I installed the latest development build on my device, Facebook would refuse to function because of permission problems. iOS was seeing it as a new app, and not honoring the permissions I'd previously granted on that device to Roastmaster.  Toggling the setting in preferences fixed it, and it was fine after that. I eventually just chalked it up to an iOS bug, or one of the slew of quirky things that happen with permissions when running a development version, thinking it wouldn't happen in a release version.

Hopefully this will fix your issues. If it doesn't or you ever notice it again, please let me know.

Thanks
Danny

335
Support / Re: Copy Roast and Default Roast
« on: December 14, 2012, 02:19:28 AM »
You and me both - every time I do back to back batches I kick myself for not doing it sooner! :) Thanks again for the impetus. Glad you like it!

336
Support / Re: Recent Roastmaster Upgrade
« on: November 24, 2012, 04:25:07 PM »
That's right. I'm going to try and get the data sheet finished and on the homepage tomorrow.

In the meantime, here's links...
  • A Phidget Temperature Sensor - EITHER:
  • A Host - EITHER:
  • The Thermocouple(s)

    Note: If you already have a WiFi laptop, you can always start small, and just purchase a temperature sensor and thermocouple. The SBC2 can be added later if you decide you don't want to have to lug the laptop out every time you roast.

    Also, I'd recommend the acrylic enclosures for any hardware you buy (except of course the WiFi adapter). Phidgets are just bare circuit boards. The acrylic enclosures provide solid protection for the Phidgets, and a good way to mount them if you desire.

    Danny

337
Support / Re: OK, I'm lost
« on: November 23, 2012, 04:43:51 AM »
Yes - I've staved off my Tryptophan turkey coma about as long as I can tonight, though the Sumatra and French Press I took to mom's helped with that! :) Lots to be thankful for. Happy Thanksgiving to you as well!!!

Let me know whenever you have questions.

338
Support / Re: Recent Roastmaster Upgrade
« on: November 22, 2012, 06:04:41 PM »
Hi Jerry

Great to hear you've hit 150!! Thanks for the kind words.

I've attached a PDF of both of the supported configurations of Phidgets hardware for Roastmaster's Data Logging, as well as a picture of my Phidgets setup. I purchased the SBC2, 4 port Temperature Sensor, WiFi Adapter, and the Acrylic enclosures for both of the boards. I mounted them to a piece of scrap wood, put a hook for wall hanging, connected the cables, and was done.

There are instructions in the help files inside the app, and diagrams in the Data Logging purchase screen, but they are very basic, and lack pictures. I'm still working on a pictorial instruction sheet to post here on the site, but the attached PDF will give you an overview of how it works.

Here's the basics. There is no formal USB in iOS, so all connectivity has to happen over an Ethernet network (via a WiFi router). That's where Phidgets excel.
  • You connect a K Type Thermocouple to the Phidget Temperature Sensor, which takes the readings from your roaster.
  • The Phidget Temperature Sensor is then attached to a host (via a USB cable), which broadcasts those readings over the network so that Roastmaster can use them. The host simply provides network connectivity to your WiFi network, and can be either the Phidget SBC2, or a laptop.
  • Roastmaster, then, communicates with the host to grab the readings from the Temperature Sensor connected to it.
Phidgets are administered from a Mac Preference Pane, or Windows Control Panel. These are very easy to install from the installers available on Phidgets.com. You do not have to know a bit of Linux - all settings are made via the software. It only takes about 5 minutes to configure via a desktop computer - most of that time is waiting for the Preference Pane or Control Panel to install.

If you don't have a laptop that can be placed next to the Roaster, I'd recommend using the Temperature Sensor with an SBC2 as a host. It costs more, but provides a portable solution. The SBC2 comes with a power supply.

Like I say, I will be putting formal instructions up soon - but this will give you an idea. Please let me know of any more questions you have. I love the Phidgets platform, though I'm sure a lot of folks would prefer a pre-packaged solution. Unfortunately, there really aren't any viable contenders that I'm aware of for iOS yet. But, don't let the do-it-yourself nature of Phidgets scare you, they are amazing little devices, and perform beautifully.

Cheers
Danny

339
Wish List / Re: Gas
« on: November 20, 2012, 01:40:34 AM »
I'm so sorry for the delay in responding. I was on vacation last week as the next Roastmaster update was in review with Apple. It's live as of about 15 minutes ago. :) And the week before, well... that was an insane week getting the code prepped. Most updates are a lot easier than this one was, but due to the nature of data-logging, and a few of the features, virtually every area of the app changed. But... I am so excited about what this one is bringing.

Thanks for the suggestion. No I don't think it's too much. Two other users asked about having a maintenance log - beyond the current basic cleaning count - something more definable. I think this might work hand-in-hand with that.

I've added this to that backlog ticket. I try to work on things in order based on popularity, but I keep track of every request.

Thanks for the suggestion!

Danny

340
Support / Re: Cheers! from Oregon
« on: November 20, 2012, 01:29:45 AM »
Hi Andy

I've been working on this update since the spring, as SO many folks have been asking for automated data logging. The number 2 request is tutorial videos. That is my only focus at the moment. I've had to hold myself back during development for the past 6 months - I've been itching to make some. Now that data logging is live, I'm excited to start.

Keep an eye on the site. I plan to have some out soon. Introductory, Data Logging, Profiles and Programs and more.

Thanks for the kind words. Please let me know if you have any questions in the mean time. I can always help.

Danny

341
Support / Re: OK, I'm lost
« on: November 20, 2012, 01:25:39 AM »
Hi Andy

Sorry for the delay. I went on vacation as the next Roastmaster update was in review with Apple. It's live as of about 10 minutes ago.  :)

To enter temperature readings from a thermometer or probe, you'll need to create a Reading curve. In a new roast, tap the blue button near the roasting gauge, and tap "New Curve". Set it to Reading, and the appropriate measurement system (C or F), and save. Then, back in the roasting console, make sure that new curve is selected (tap the graph to toggle through each curve in the roast).

As the roast is in progress, all you need to do to enter temps is tap the digital readout near the gauge, and enter the temp. It will create a timestamped node at that temp. You can define as many curves as you need for recording temps.

Once you hit on a good one you want to match for future roasts, you can copy and paste it from the roast to a new profile or program as a target to match in future roasts. Whenever you select that profile or program in a roast, the target curve will be visible in the roast as well as any new curves you define in that roast for recording new temps.

Hope this helps. Let me know if you have any other questions.

342
Roasting / Re: Roast Number 400!
« on: October 29, 2012, 07:37:03 PM »
Ahhh - you're catching up to me... :)

Thanks for posting this!


343
Wish List / Re: Export data
« on: October 23, 2012, 10:14:30 PM »
Hi

Sorry for the delay in responding - I was wrestling this weekend with a bug I found testing the next update. It was a rarely presenting bug with the new data logging option, but in need of fixing nonetheless.

There are no docs for the database schema at the moment, but enhancing the import/export is something I want to tackle soon. I try to go by popularity of requests, and that is one of the main ones after some the new features in the next update. Roasts are certainly the main request, but I can see the advantage of exporting blends as well.

In the mean time, you'd have to do the querying with SQL. The ZBLEND table uses a join table for its to-many relationships to beans and roasts - it's called ZBLENDEDITEM. I'm not sure how proficient you are with SQL, or if you have software to open SQLite databases, but a stripped down query would look something like...

select b.ZNAME, b.ZNOTES, bi.ZAMOUNT from ZBLEND b join ZBLENDEDITEM bi on bi.ZBLEND = b.Z_PK

That would just give you rows with blend name, notes and amount for each blended item. It gets complicated from there, because Roastmaster lets you blend either a roast or a bean, so to get anything useful, you'd have to join another table in based on whether it's a bean or a roast that you've blended - based on whichever column holds data out of ZBEAN or ZROAST. It can get complicated fast, especially considering recursion if you blend a roast of a blend of a roast of a... etc., so a home-rolled query may prove very difficult. But this is something the app already does for you behind the scenes, so adding support for blend export is not a monumental undertaking, which is why I'm considering it for roasts and blends.

DISCLAIMER : please don't ever modify data in any way outside of Roastmaster. Core Data on iOS is VERY finicky, and will refuse to work if even the smallest of database details is not in adherence to its own standards.

Sorry - this may not have been helpful to you. But, if you have a sample sheet, even with dummy data - I'd love to see an excel file to help me get a good picture of what you're after. Feel free to email to support@rainfroginc.com.

Cheers
Danny

344
Wish List / Re: A video tutorial and a desktop version would be nice
« on: October 23, 2012, 09:40:23 PM »
Hi

Thanks for the kind words. Sorry for the delay in responding, I'm eating and breathing code at the moment with the next update. :)

Tutorials are sorely needed, and I promise they're next on my to-do list. I try to keep myself on track with what the majority of folks are asking for, and you've hit the big one on the head - data logging. I'm wrapping up the next update as we speak, and the data logging option is included in the code as an add-on. It's been a huge undertaking, as it permeates into many areas of the app at the lower levels, but it's finally almost ready.

Sorry to those who have heard me say that 4 weeks ago - the release of the new iPhone with its new armv7s processor, and iOS 6 have both thrown a monkey-wrench in the works, and added a number of issues I wanted fixed before the release. I knew when starting data logging, that it would essentially black-out any maintenance updates. They wouldn't be possible until data logging was done. Some long overdue improvements will be included - iPad retina, iPhone 5 retina, and a slew of others.

You are correct - external hardware support in iOS is VERY sticky, and the MFI program is dark cave I fear to tread in, even if there were viable options on the market for bluetooth. Ethernet, though, is another story. I've chosen an open-source hardware to support for now that is ethernet based. I've been roasting with it for 6 months with Roastmaster, and it has proven extremely reliable and more cost-effective than other wireless options. Other devices can be added as I evaluate them for inclusion in the future.

I've spoken with Apple about this, and was informed that they don't tend to regulate ethernet connectivity in apps, so hopefully the update will pass through without incident. They warned me, though, that they will not give preliminary approval on anything. For that reason, I have to keep it a bit quiet, in case there is a problem in app review. I will post a full write-up on the hardware as soon as the update is approved.

I don't have a desktop version planned at the moment - though a number of folks have asked. I don't truthfully know how much appeal that would have, outside of the data logging issue. To me, it's much easier having an iPad by the roaster - and if that feature can be supported on iOS, seems the desktop version may not be an issue any more.

If you have samples of logs, feel free to them to me - support@rainfroginc.com. The next thing I want to work on is import/export abilities. Lots of folks are asking for enhancements in those areas. Would love to see exactly what people want to achieve and what kind of data they are working with.

Cheers
Danny

345
Support / Re: Connection thermometer or thermocouple to device via USB
« on: October 16, 2012, 01:49:01 AM »
Hi

Unfortunately, no. iOS doesn't offer true USB support, so Apple's implementation of USB for cameras via the adapter is a low-level thing not exposed to developers. In other words, all of the necessary USB protocol handling that exists on desktops, doesn't exist on iOS. It's possible to make a few things happen via that adapter, but only if code is exchanged in a protocol that emulates a keyboard, but that doesn't help with USB probes.

The good news is, that the next update will introduce a Data Logging option for one particular brand of open source hardware. I was code complete a few weeks ago, right before iPhone 5 hit the market, and ready to release. Once the phone hit, though, I decided to delay the update for a few weeks while I added support for the new retina screen, and ironed out a glitch with the iPhone's new Arm7s processor and the data logging hardware.

I am hoping to have the update to Apple within a week. I think I can finally say that with certainty - as nothing new iOS-wise is coming down the pike for a while to surprise me. I'll post a full write up on the blog with a link on the home page once the update is approved. From the conversations I've had with Apple, what I'm doing is not frowned upon, because it functions via Ethernet, but they won't promise approval until the binary is submitted for review. Otherwise, I'd post the details so folks could get the hardware now in the meantime. As it stands, it's just a bit too risky - I don't want people stuck with hardware if Apple decides to say no.

Cheers
Danny

Pages: 1 ... 21 22 [23] 24 25 26