Gene Cafe CBR-101 vs Hottop KN-8828D/B/P
7 May, 2008
Contents
- Introduction
- First Impressions
- Operation
- Roast Tests
- Observations In Use:
- Cupping
- What's a Corretto??
- Programmable Hottops - B and P models
- Summary
- Conclusion
- At A Glance List Of Pros And Cons
Introduction
First things first, this is not an exhaustive review going through every little feature of each machine as there are other reviews around for that. This is more a side by side of the two machines, picking up some of the features of each machine and comparing the appearance and taste of a few roasts, which ultimately is what matters.
When you've got yourself a fancy espresso machine, a good grinder and a nice set of accessories, what's the next step for a coffee afficionado? Well for many, it's the venture into home roasting. While a reliable supplier of commercially roasted beans is the choice of many, you can't get much fresher than roasting yourself, and the big plus of using green beans is shelf life - roasted beans should be used within a few weeks of roast whereas greens can last for several years under the right conditions; consequently some home roasters have a considerable cache of green beans patiently waiting in a cellar, the back of a cupboard or in the back shed for the right time to hit the roaster. Beans, their origins and how to roast them is an enormous field, perhaps the largest sub-discipline of the coffee world simply due to the number of permutations of bean varieties and how they react to different roast times, temperatures and temperature profiles; perhaps this is some of the attraction as home roasters start to understand the importance of these different factors and try their hand at blending.
A 'Corretto' home roaster, made from a heat gun and bread maker. Courtesy coffeesnobs.com.au
There's been a real grass-roots approach to home roasting for some time. Inventive coffee lovers have used popcorn makers, heat guns, bread makers, barbecues, ovens, hotplates and even cobbled their own contraption together, all with the aim of consistently and repeatably turning green beans into roasted beans. You have to admire the inventiveness of some people and the lateral thinking required that can suddenly turn innocent equipment in a supermarket, homewares store or appliance store into implements for roasting coffee! We'll have some interesting comments in our conclusion on the effectiveness of these devices.
Effective as these devices may be, however, some like their roaster to look the part rather than being clearly putting objects to a secondary purpose, and to have easy-to-use integrated controls with a good fit and finish, something often hard to achieve for a backyarder. For these people there are commercially available home roasters. These devices are normally based on the same two fundamental methods of roasting as most home contraptions, where either heated air is blown upwards through the beans to cook and agitate them (fluid bed roasters such as the iRoast and Imex units), or where a cylindrical drum containing the beans is heated and rotated to cook and agitate the beans (drum roasters such as the Gene Cafe and the Hottop).
This is unashamedly not the first comparison of these roasters. There are a few already circulating the internet, however reviews and such are useful only to a point in that they ultimately reflect the opinion of the person(s) writing the reivew. I found fairly consistently that the Gene Cafe is reported to have a brighter, livelier roast where the Hottop is more full-bodied but comparatively flat, and the Gene often wins the 'competition' on that basis. However, that appeared to be based on the premise that a bright roast is what's desired - would the Hottop perhaps be a better choice for someone who prefers a more full-bodied roast? Is it a universal law that the Gene cannot produce full-bodied roasts or that the Hottop cannot produce bright roasts? Without knowing the personal preferences of the reviewer it's hard to know. A colleague of mine had purchased a second-hand Hottop, so I picked up a new Gene and we got together for a bit of a comparison!
Before we launch into the review proper, it's appropriate to mention where these units under test fit into the model range. The Hottop is made by Taiwan's Chang Yue Industrial Corp, and the KN-8828D is the second model produced by Chang Yue. There was an analog model initially with only LED readouts, then came the D model which has a digital display for time. Both these units have been on the market for a while, but what they both lack is programmability. In 2007 Hottop released the KN-8828B and KN-8828P models, both of which are programmable and we'll touch on those briefly at the end of this review. The Gene Cafe comes from Genesis Company in Korea and, as the model number suggests (CBR-101) it's the first release of the unit; however it's completely programmable in terms of time and temperature.
So we're not exactly looking at an apples with apples comparison here - one is a unit with about five years maturity in the market and no programmability while the other is relatively a newcomer with all the potential teething problems inherent with that, but with the benefit of programmability. It might seem an unfair match in some ways, however these models were chosen because at the time of testing they were the units most readily available to Australian consumers, the ones which most were comparing for purchase and the most comparable in price. The updated B and P controllers are touched on later in this review.
First Impressions
My first impression when I saw the box the Gene came in was 'Wow that's big!'. From photos I was somehow expecting something about the size of a shoebox, but I was wrong! The Hottop is a bit larger but is more the size I expected somehow but neither of them are miniature by any means. Facing the control panel, the Hottop stands 35cm tall, 48cm wide and 25cm deep, so it's no miniature (for those who own a Rancilio Silvia, the Hottop is the same height, but wider and not as deep). It weighs about 10kg. Including the chaff collector, the Gene is 23cm tall, 49cm wide and 25cm deep and weights 5.5kg. So they're basically the same size except the Hottop is taller and heavier. Both units are solidly made and well designed and both appear to be good quality units. However the Hottop is predominantly metal with a cooling tray and looks more like a minituarised proper roaster, whereas the Gene Cafe is mainly plastic and looks (to the untrained eye) a bit more like a toy. So the Hottop more looks the part even if only because it's mainly metal (it has its own design quirks which let it down).
Operation
The heating element in action on the Hottop
The Hottop, as described earlier, is a traditional drum roaster. Looking at it end-on at the viewing window, it has a heating element on the left side of the unit between the outer shell and the drum, and the drum rotates anticlockwise. This means the drum is heated by the element, while the beans sit roughly opposite the heat source as the drum rotates and pulls them up the right side of the drum. The drum also has a spiral to move the beans towards the rear of the unit, at which point they are pushed upwards to about the mid point of the drum and are pushed back to the front of the drum by the beans moving along below them. It's quite a good system and means the beans are always on the move and never get too close to the heat. Speaking of heat, this particular version of the Hottop has just one heating profile - the only control you have is how long the roast goes for (which is limited to 23 minutes) and a Eject button which immediately dumps the beans into the cooling tray. The newer versions of the Hottop are more programmable.
The Gene is quite different. It has a glass drum containing a stainless steel skeleton, the combination of which contains the beans. At each end is a vent, and this carafe sits off-centre in the roaster. This means as the drum rotates, the beans are moved not only round and round but from end to end, providing very good bean agitation. The heat source comes from a heating element in one end of the roaster which blows hot air in one end of the drum, onto the beans and stainless steel skeleton, out the other end of the drum and into the chaff collector. While it may sound like a fluid bed roaster, the Gene is still a drum roaster as the hot air is only used for heating the beans, not for agitating them. The control system is quite different to the Hottop. The user selects the roast time and temperature and the machine goes for that. At any time you can increase or decrease the temperature or roast time.
The only anomaly about the Gene's control panel is that the countdown time is shown in tenths of a minute, so each least significant digit represents six seconds - from 18.0 you go down to 17.9 not 17.5. In decimal terms this makes sense and I guess it's more accurate having this last digit represent six seconds than ten, but since a minute is 60 seconds not 100 it takes a bit of getting used to thinking of minutes in base 100 rather than base 60. Otherwise it can be tricky trying to convert 17.8 minutes to 17 minutes, 48 seconds on the fly!
Airflow paths on the Gene and Hottop
The Hottop employs a suction airflow system where the fan sits at the end of the line and sucks clean air (and roasting fumes) through the roasting chamber from a vent above the drum. This system demands a filter to remove the smoke particles before they reach the fan which is a minimal but ongoing expense (Hottop recommend changing the main filter every 20 roasts). Air flow in the Hottop is fairly minimal as the fan normally only comes on towards the end of the roast though this can be controlled in newer versions. The Gene by contrast is a compression system where the fan sits at the start of the line drawing clean air from outside and pumping it through the roasting chamber, collecting the chaff which is then deposited in the chaff collector and venting the fumes to atmosphere. On the positive side, the Gene's system deals with the chaff and fumes in one fell swoop and eliminates the need for filters, but on the negative side the increased airflow makes it a noisier unit and makes it harder to hear the beans cracking during roasting.
So both roasters aim to do the same thing but by fairly different methods. The Hottop has a potentially 'set and forget' system with its one profile, whereas the Gene gives you full control over what the roast is going, though any profiling has to be done manually. How did they go?
Roast Tests
We decided to do three 250g roasts. The first one we would use the machines exactly as designed - the Hottop on its standard profile and the Gene set static at 228 deg C. Both roasts would be stopped at around the start of second crack and cooled using the roaster's provided cooling system. The second roast would be the same on the Hottop but since the Gene takes a long time to cool normally, we would hit the emergency stop once second crack started and cool the beans externally. The third roast we would endeavour to manually track the Hottop's temperature profile on the Gene so the resultant roast should be comparing purely roast process rather than time and temperature; the Hottop's chaff tray was opened slightly at the start of first crack to extend the roast a bit during this period.
Roast 1 - 250g Ethiopian Harrar
Both roasters were left on default settings - the Hottop followed its standard profile, and the Gene was set at 228C.
| Parameter | Hottop | Gene | Notes |
|---|---|---|---|
| Ambient Temp | 29C | 34C | Hottop pre-heat time 4:40, pre-heated temp 60C |
| First Crack | 16:30 | 12:30 | HT temp 197C (195C bean) |
| Second Crack | 19:00 | 18:00? | Gene 2C unheard, based on colour |
| Roasted Bean Mass | 217g | 212g | |
| CS card full colour range | 5-11 | 6-11 | Only a few samples of each at the extremes, particularly the light end of the scale |
| CS card median colour range | 8-9 | 9.5-10.5 | |
| Number of beans with tipping | 3 | 9 | (after a couple of minutes of similar scrutiny) |
Other notes:
While the Gene was stopped when the roast looked about the same as the Hottop's, its cooling cycle took about another 12 minutes to bring the temperature down from 228C to 60C. Beans from each roast were opened and noted to be even throughout.
On this first roast we also recorded the environmental temperature of both roasters and the bean mass temperature on the Hottop. The Hottop's design makes it easy to add a bean mass temperature probe to see what the beans themselves are doing and we've plotted these below. It's useful to know this sort of information for the Hottop as, providing you take the roast progress into account, you could potentially do away with a bean mass probe as you'll know roughly what temperature the beans will be at for any given environmental temperature.
| Time (mins) | Temp (environmental) | Temp (bean mass) | Gene (environmental) |
| 0 | 60 | 60 | 30 |
| 1 | 80 | 75 | 118 |
| 2 | 93 | 90 | 160 |
| 3 | 104 | 99 | 184 |
| 4 | 117 | 109 | 200 |
| 5 | 127 | 118 | 213 |
| 6 | 136 | 124 | 224 |
| 7 | 145 | 133 | 228 |
| 8 | 154 | 143 | 228 |
| 9 | 162 | 151 | 228 |
| 10 | 169 | 156 | 228 |
| 11 | 174 | 160 | 228 |
| 12 | 178 | 163 | 228 |
| 13 | 182 | 168 | 228 |
| 14 | 187 | 176 | 228 |
| 15 | 191 | 183 | 228 |
| 16 | 197 | 193 | 228 |
| 17 | 204 | 203 | 228 |
| 18 | 208 | 209 | 228 |
| 19 | 211 | 217 | 228 |
Temperature profiles of the Gene and Hottop
Roast 2 - 250g Indian Monsooned Malabar
The Hottop ran the same cycle as roast 1, however with the Gene we stopped the roast with the emergency stop and manually cooled the beans out of the roasting chamber. Interestingly the beans were still quite warm after several minutes whereas the Hottop's fan-forced cooling tray + stirrer got the job done within a minute or two.
| Parameter | Hottop | Gene | Notes |
|---|---|---|---|
| Ambient Temp (degrees C) | 40 | 42 |
|
| First Crack (m:s) | 15:55 | 9:40 | HT temp 199C (bean temp 193C) |
| Second Crack (m:s) | 19:00 | 11:00 | HT temp 212C (bean temp 218C) |
| Roast Stop (m:s) | 19:10 | 11:40 | |
| Roasted Bean Mass (g) | 223 | 212 | |
| CS card full colour range | 6-11 | 5.5-10.5 | |
| CS card median colour range | 9-10 | 9-10 | |
| Number of beans with tipping | 0 | 0 | 3 of each had small scorch marks |
After the Hottop had done its 30 min cooling cycle (during which we ran the Gene's tests), we started round 2.
Roast 3 - 250g Ethiopian Harrar
The Hottop's chaff tray was opened a small amount at the start of first crack to extend the time to second crack a little. With the Gene we manually tracked the profile used by the Hottop in roast 1 to give a ramped temperature roast rather than a 'high-pass filter' profile. This allows us to compare purely the effect of the different roasting methods rather than time and temperature. The difficulty was that we weren't able to take any bean temperature measurements on the Gene, so unlike the Hottop we didn't know the actual difference between the bean temperature and the environmental temperature. Initially we took a guess that the bean temperature would be about 10C less than the environmental temperature as applied with the Hottop so all the Hottop's environmental readings were increased by 10C and manually updated on the Gene every minute. This however turned out to be an underestimation as first crack didn't come until 22 minutes even though we'd cranked the temperature up markedly at the 20 minute mark. So we did a second roast with a loading of +20C which is the roast used in the table below as this was a much closer match.
| Parameter | Hottop | Gene | Notes |
|---|---|---|---|
| Ambient Temp (degrees C) | 42 | 42 | |
| First Crack (m:s) | 18:00 | 18:00 | HT temp 202C. Gene temp 232C |
| Second Crack (m:s) | 20:50 | 21:00 | HT temp 214C. Gene temp 244C |
| Roast Stop (m:s) | 21:15 | 21:00 | HT temp 215C |
| Roasted Bean Mass (g) | 210 | 216 | |
| CS card full colour range | 6-10 | 6-10 | |
| CS card median colour range | 9-10 | 9.5-10.5 | |
| Number of beans with tipping | n/r | n/r |
Notes
The Gene would take about 15 seconds to increase its temp from one reading to the next so it's not an exact ramp but close enough. Also this was a very chaffy roast. Also the slower Gene roast seemed to have more chaff than the faster one
Observations In Use
Some general observations about each machine as a result of the roasting tests above.
Noise
Gene Cafe - Roasting
Hottop - Roasting
The Hottop is noticeably quieter than the Gene for several reasons. As mentioned, the Gene uses a fan to blow hot air through the machine as well as a noisy motor to rotate the drum, whereas the Hottop has a passive heating element and a quiet drum motor (it has a fan as well but this only comes on occasionally and is barely noticeable when it does). The bean agitation is also more intrusive on the Gene than the Hottop (discussed shortly); this is probably because the Hottop keeps the beans gently moving by the same amount all the time, whereas on the Gene the beans are normally sliding around a bit but there's also a point where they don't move at all and another where they're crashing from one end to another - so the noise is both greater in volume and less consistent, both of which make it harder to filter out than on the Hottop. Thus the main sound of the Gene is a combination of the whiney motor, the regular crash of beans from end to end of the drum, the knock of the chaff sweeper against the drum and the dull roar of the fan. The Hottop is mainly the tinkle of beans and a quiet graunching sound as the drum rotates. Because of this, it's much easier to hear the bean cracks in the Hottop. You hear them without listening, whereas the Gene you really have to train your ear to pick the cracks over the noise of its operation. You can apparently hear them more easily if you listen at the chaff collector, however bearing in mind this is breathing out 200+ degree C air it's a slightly risky exercise! The video further down of cooling in the Hottop has a section near the start where you can hear second crack starting. The roar in the background is the fan on its fastest setting and it's still easy to hear the cracks.
Chaff and Smoke
Gene Cafe chaff collector
The Gene deals with chaff quite well. Chaff from the beans initially is blown to the exit end of the drum where it's supposed to pass through a coarse mesh. Invariably some gets stuck, so as the drum rotates, a weighted wiper falls from side to side across the mesh to wipe off any chaff which tried to go through sideways and it gets another go. It doesn't work perfectly but it's about the best it could be as you'd have beans escaping if the mesh was much larger. Once the chaff's out it falls into the bottom of the chaff collector while the hot air exits through a fine mesh in its top. Once the roast has finished and the chaff collector cooled, it can be opened to clean out the chaff. All in all it's a very neat and well-designed system; little chaff is left in the chamber and everything that's blown out is in one easy place to empty.
Hottop chaff collector
The Hottop doesn't do as well generally - it relies on chaff falling through the circular holes in the drum and into the chaff tray below the drum. It's a simpler idea but the holes really aren't big enough to do it properly - it would be better if there were also some long slits to give some other options for the chaff to get out. The result is more chaff remaining with the beans, though what does escape is relatively easy to remove via the chaff tray.
Build Quality
Air gap between the control panel and the body of the Hottop
Chaff left behind in the Hottop's body after removing the drum
The Hottop certainly looks the better-made of the two units - it's predominantly metal and feels good quality whereas the Gene is plastic - nice plastic, but still plastic. However the components of the Hottop don't fit too well together. This isn't immediately obvious when you look at the machine, but once you use it you discover smoke wisping from all sorts of unlikely places - around the front of the drum itself, the bean entry chute, I've even seen some coming out from behind the control panel! This also means that chaff wends its way into places it doesn't belong, and once in a while you'll need to remove the back of the machine to blow out the stray chaff. It's quite common to get a nice little collection underneath the chaff tray as well. The Gene may not look as upmarket but fit and finish appears to be a bit better. This is probably a benefit of being designed more recently, where the Hottop's fundamental design is unchanged from 2002.
Bean Agitation
A common place for beans to get stuck in the Hottop
An extreme example with some smaller beans of what can happen with the Hottop
As noted earlier and in the roasting videos, the two roasters use quite different processes to keep the beans moving. The Hottop's drum rotates at about 48rpm and employs a steel wormwork around the inner wall which drags the beans to the back of the drum where they then work their way back to the front and the cycle recommences. The rest of the drum, apart from structural supports, is open space. The Gene, again, is quite different. It rotates at just 10rpm, but the off-axis positioning of the drum provides extra agitation so the slow speed isn't as bad as it reads on paper (or screen!). The drum itself has a wide steel divider running right through middle of the chamber, and during rotation this lifts the beans up above the vertical centre of the drum and causes them to slide across it and crash onto the lower side of the drum - so there's no way any of the beans can stay still. There's normally a second or so where the beans are stationary before they crash across to the other side / end of the drum but the amount of agitation is excellent and probably more than the Hottop.
On the Hottop, invariably one or two beans get stuck between the wormwork and one of the holes in the drum and just gets cooked, and looking through the small glass window at the front some beans seem to sit there for quite a long time. However as the results show, this didn't particularly affect the evenness of the roasts either way, probably because the whole chamber is about the same temperature and in both cases the beans are away from the direct source of heat (apart from any stuck beans in the Hottop). On the Gene, you can't fault the agitation, but the downside to this extra movement is the noise the beans make in moving which has already been discussed.
Temperature Sensors
Hottop internals - heating element (left), vent (top front), bean entry chute with aftermarket temperature probe (top rear), bean exit chute (bottom rear)
Both roasters use an environmental temperature sensor, meaning they measure the temperature of the air going in or out of the roaster rather than the actual bean mass temperature. The latter is a far better indication of what's going on but is obviously more difficult to achieve. The design of the Hottop makes this a much easier aftermarket addition, and in fact the unit on test had such a modification which incorporated a bean mass temperature probe and this showed the environmental reading was about 10-15 degrees C higher than the bean sensor up to first crack, after which the bean temperature rose above the environmental by about 5 degrees C. The Hottop's design makes it fairly easy to insert a probe right into the bean mass, via the bean entry channel for those keen to do a bit of drilling. Without ruling anything out, the design of the Gene makes this basically impossible - not only is the whole drum removed for filling, the drum has a large stainless steel divider which would hit the probe twice per revolution, and in any case the beans are tumbling from end to end of the drum so the probe would be out of the bean mass for about 50% of the time.
Design Niceities
Both machines have some nice design features. The Hottop's fan comes on briefly around 100C to blow out any steam which may have been emitted by the beans, and if a bean gets stuck in the cooling tray the cooling paddle instantly reverses direction. When stopping the Gene, it slowly rotates the drum to a position where you can easily grasp the handle for simple drum removal (see cooling video)
Cooling
Hottop cooling tray
Gene Cafe - Cooling
Hottop - Cooling
Standard cooling on the Gene is very poor. You've got a hot chamber full of beans sitting at 200+ degrees C and they're expected to cool by simply blowing air through the heated chamber? As noted it took 12 minutes to drop the atmospheric temperature from 228C to 60C. Obviously it's a compromise Genesis had to make to get the design they were after but if you use factory cooling you really need to stop the roast a couple of minutes before you want it to finish which is a bit of a guessing game. Alternatively use the emergency stop feature (as we did for the last two tests) which allows you to pull the chamber out and dump the hot beans into colanders or something for external cooling. Works a lot better and a lot faster.
The Hottop by comparison does a great job. As soon as you've finished the roast you hit the Dump button and within 20 seconds all the beans are out and being fan-cooled and agitated. They were noticeably cool within a minute or two and just warm to the touch within five minutes.
Cupping
This was the big test - how did they perform in the cup? A sample from each roast was ground a week after roasting and placed in hot water with the roaster marked underneath, and we didn't know until after which roaster supplied the sample. One thing that was interesting about the blind cupping was that from the the common dogmas about these roasters, I had certain ideas about which roaster provided which sample. I was very surprised when I actually realised the truth of the matter. I had intended to write up the notes of each sample as it was tasted over time, however what's probably more useful is the overall impression, and what commonly held beliefs were challenged.
The Big Myth - pretty well the summary of what I've read of these two roasters is that "the Gene gives a brighter, more acidic finish whereas the Hottop gives a duller but more full-bodied roast"
Reality: The Hottop actually gave the brighter, more acidic finish on the Harrar and the Malabar while the Gene was brighter on the Peaberry, however the Hottop's roasts were universally more intense and more full-bodied. Our final roast was particularly interesting because we managed to mimic the Hottop's roast time which allowed us to test the heating process rather than the roasting time - did radiated heat provide a better result than heated air as I expected it would? We actually did two roasts on the Gene; the first has the set point 10 degrees above what the Hottop's bean temperature was at the same time however the roast ran for nearly 25 minutes so we repeated it with a 20 degree difference and this yielded basically exactly the same roast time as the Hottop. To the taste, the first of these roasts (the long one) was flavourless, thin and had a sharp - completely unpleasant. The second roast was better, more acidic than the Hottop had given and more citrus notes though fairly thin in body. The Hottop gave medium acid, chocolate notes and a clean aftertaste.
Overall I preferred the Hottop's Harrar, the Gene's Malabar (just) and the Hottop's Peaberry. As a milk-based drink, my wife and I initially preferred the Hottop's Harrar, the Gene's Malabar and were split between the Hottop and the first Gene roast on the Peaberry. However a few days later we found we preferred the Malabar from the Hottop and the second Gene roast from the Peaberry! Robin however preferred the Hottop's Malabar and Peaberry and the Gene's Harrar though noted this lost its brightness over the week.
What became obvious is that the temperature and heat profile had far more effect on the resultant roast than any other factor, be it the heating method, noise, chaff collector or fit and finish.
What's a Corretto??
A 'Corretto' home roaster, made from a heat gun and bread maker. Courtesy coffeesnobs.com.au
This is a slight aside but it's relevant. Coffeesnobs member corretto came up with the idea of roasting coffee beans via a heat gun with the beans sitting in a breadmaker. The breadmaker provided the bean agitation while the heat gun provides the heat. If the heat gun is on an adjustable stand of some description, the heat can be finely adjusted by either moving the heat gun closer or further away, or if it has the feature, by adjusting the temperature on the heat gun. Often a digital multimeter with a temperature probe is used in conjunction with this setup which allows a relatively accurate indication of bean temperature. It's very much a grass-roots approach to roasting and one that can be easily cobbled together for less than $100 from a $40 heat gun, a $25 second-hand breadmaker and a few odds and ends; so it's understandably a solution most would see as a poor man's roaster compared to these fancy purpose-built devices, however it gives excellent results in its own right, particularly given the low cost involved.
One interesting observation Robin made was that: "The Malibar is far better in the Corretto than either HT or Gene..... Harrar is about the same in the Corretto as best I can recall (some time ago since I've roasted it).... Peaberry is best in the HT.... but only a little ahead of the Corretto. So maybe spending $600 -$1000+ isn't gaining you much - if anything, other than ease of use ;-) "
Hottop B and P controllers - programmable!
Just before this review went to press (uhh, web), I received a Hottop P model as well as a B model upgrade kit. These are really the subject for a separate review but I'll mention them briefly here because they change the landscape significantly. Both controllers give you what the D controller lacked - programmability, so immediately the stage is set for a showdown with the Gene. Comparing the Gene to the Hottop D yielded mixed results; from my initial testing of six roasts with each of these controllers, both Hottop controllers have yielded consistently more pleasant aromas and flavours than the Gene. That's not a hard and fast conclusion but it's a very positive initial sign for the new Hottop controllers.
B controller:
Hottop roaster fitted with 'B' model controller
Hottop 'B' model controller in use
The B controller is similar to the Gene's system in that you enter a roast time and a target temperature and start it running. At any time throughout the roast you can adjust the heater power in increments of 10%, and the fan speed in increments of 25%. Once you've finished the roast you can save the profile you used in one of three memory locations. One criticism I had read of this controller was that you didn't have very long to save the program before the option disappeared; on my controller it waits about three minutes for you to save the program which I think is more than sufficient time. The B controller has everything the Gene does with the addition of a memory and a relatively accurate drum temperature sensor.
On the downside, there's no way to modify or review a program before using it - you can only save or overwrite programs by re-roasting with the program and making the changes as you go; which can be annoying because if, for example, you brought the fan on too early, you'd have to watch for when it comes on and then instantly turn it off again, so the program then becomes 'fan on for one second then off'. The irritation with control by time (B model) rather than time (P model) is that in a given roasting session, the temperature in the first roast normally rises faster during the first five to eight minutes than during subsequent roasts. This means that your decisions in the first roast of when to, say, drop the heater power or bring the fan on, may come too early in subsequent roasts which sets up the aforementioned condition of a messy program. You also have to learn how long it takes for changes in control panel settings to take effect in the drum - I stalled two of my first four roasts by dropping the power too much when I was trying to slow it down. This is however no more or less of an issue than with the Gene, it's just that you don't know with the Gene what's actually going on inside the drum! It's also a pity the B controller only has three program memory locations compared to the P's nine, and that these are not renameable - AD1, AD2 and AD3 mean nothing to me.
P controller:
Hottop roaster fitted with 'P' model controller
Hottop 'P' model controller in use
The P controller is a different beast altogether. Rather than starting the roast and then tweaking things as you go, the P allows you to program a profile before starting the roast. Programming is done in eight segments each of which can be between one second and three minutes long, within each of which you set a target temperature and fan speed. This might seem a bit restrictive but the range available within each segment is quite adequate. The initial program does take a long time to enter, however once you've done it it's easy to copy and paste the program to any of the available nine memory locations, each of which can be named. As you roast the only adjustment you can make is that of the temperature in the active segment. As with other models, you can add 'plus time' if you want the roast to go for a bit longer, and dump the beans once you're finished. It's also the only Hottop control panel (thus far) that doesn't consist of buttons behind a flexible plastic film. The buttons feel like silicon and are very nice to use. The illumination on the control panel changes depending on the mode you're in (roasting, programming, erasing and so on). I've only done one roasting session with this controller but I loved it - I took notes on how things went, cloned the program to a spare memory location, adjusted the temperatures and times as I felt fit and started again. Since the P controller's programs are temperature-based rather than time-based, the system will control the heater power to aim for the temperature you've specified in the time you've specified regardless of any other variables. Compared to the B controller, I find this works better in practice even though prior to use I preferred the concept of the B controller. To use the same example as in the B section above, if a subsequent roast is taking longer than an earlier one, the P controller will keep applying the power until it hits the temperature you want, rather than backing off the power at a given time when such a decision may not be warranted. I also prefer being able to decide what I want the roast to do before starting rather than having to make major decisions on-the-fly if the temperature is getting too high or too low as you have to with the B controller.
Negatives of the P controller - well it certainly takes a long time to do your initial program, that's undeniable but fortunately it's something you only need to do once. In general it isn't as flexible as the B controller once the roast is underway - you really need to decide what you want the roast to do before starting. Once you're roasting the only thing you can adjust is the temperature of the active segment - the temperature in pending segments can only be changed once the roast moves to those segments and you can't adjust the fan anywhere, so if you find the chamber's getting too smokey too quickly there's nothing you can do about it. Another irritation is that while each segment is time limited to three minutes, you can't regain in later segments any time removed from earlier ones. For example, while a maximum roast time is 24 minutes (eight segments x three minutes each), if you made segments three and four only one minute each, your roast now cannot exceed 22 minutes. This isn't that likely to cause a problem so long as you don't reduce segment time too much but if you cut out too much time early on you may find your program running short. The idea of choosing a temperature and letting the roaster take care of the rest is good, but it only appears to look at the active segment rather than looking ahead to what's coming and plan for it - so with 10 seconds to go if the temperature is too high it'll drop the heating power even if the next segment will require the power to come back on. The nine memory locations are good, but (perhaps being picky) it would be nice to be able to use more than three characters for a program name; in some cases this is sufficient but I'd like the flexibility to go further so if I have three different profiles for a given bean I have more than two characters available for the bean type. During roasting, the word 'roas-ting' appears across two lines so there's no reason why program names up to eight characters shouldn't be available even if you don't want to use all eight characters.
Summary
Gene Cafe
Compared to the Hottop it's much faster to use. Once you set it up, you can place the drum stand and the drum on a set of scales, zero the scales, pour in your 250g of beans, insert the drum into the roaster and start. It's a process that might take 60 seconds if you're careful about it, it can be done faster. The control panel is simple to use and relatively useful during roasting (though the base-10 minute digit takes a bit of getting used to) and it's nice you can adjust time or temperature at any point in the roast. It's really easy to see the beans, it deals with chaff really well and roasts as evenly as the Hottop. Back to back roasts are nice and quick too, there's no ongoing cost in replacable filters, there's a proper Australian distributor which means warranties are easily dealt with (edit July 2008: Hottop is now supported in Australia so the Gene isn't unique in this regard anymore), and at RRP $AUD780, the Gene certainly presents a compelling argument value for money.
But there's another side to the Gene. My biggest gripe (and that of many users) is the high and inconsistent level of noise it makes. Undeniably, one of the key indicators used by a coffee roaster is listening for subtle cracking sounds of the beans, and the Gene's combination of whiney motor, periodic clatter of beans, thump of the chaff wiper and roar of the fan mask all but the most enthusiastic cracks. Those with plenty of Gene experience claim they learn to hear them and I'd agree that my later roasts were easier in this regard than my first; however it's a struggle at best and impossible at worst, and the inability to do this accurately and consistently is a big downer in my books. Regardless of how good the rest of the unit may be, if the operator has difficulty identifying key roast indicators then the result may still be a spoilt roast. The plastic components lend a 'cheap' appearance to the unit, especially compared to the Hottop. Back to back roasts are quick, but the trade-off is that the starting temperature could be anything from ambient to 150 degrees C so repeatability could be a problem. The temperature readout bears little correlation to the temperature of the beans; after seven minutes the roaster may be producing 230 degree C air but the beans wouldn't be anywhere near that, which makes any attempt to roast to a 'profile' difficult at best. This is exacerbated by the fact that the otherwise good design of the Gene prevents the installation of a bean mass temperature probe to give an accurate indication of what's going on with the beans themselves. Cooling is a real pain too, and the only way to effectively cool is to utilise the emergency stop feature, remove the drum without burning yourself and dump the beans into some external cooling device. All these points one could probably adjust to over time but it requires a long and potentially frustrating learning curve, especially if you're comparing it to the Hottop.
Hottop
The Hottop certainly looks the part, and it's QUIET! The most valuable tool to a roaster, the ear, has ample scope for use here. It's not difficult to hear cracks standing next to the roaster or even from the other side of the room, which is impossible on the Gene. It provides a nice gentle heating profile which attempts to emulate a traditional drum roaster. The cooling system on the Hottop may not be as quick as a commercial system but it leaves the Gene for dead. Beans are quite touchable within two minutes of the roast finishing, something the Gene can never achieve. Two criticisms often levelled at the Hottop are the need for replacable filters and that it's slow for back to back roasts; I actually feel both these points are there for a reason.
- Firstly the filters - yep it may be a pain to have to shell out a few dollars on filters every 50 or so roasts, but you only have to see the comparatively exorbitant amount of smoke produced by the Gene and the amount of smoke which comes off the Hottop's beans when they're dumped to see why the Hottop has a filter. Apart from the smoke that escapes from its poor seals (which we'll come to in a minute), most of the smoke particles are trapped by the filter which is a plus for the operator (and possibly for the environment, 'possibly' because the filter needs to be disposed of eventually).
- Yes it's slow for back to back roasts but that's for a reason. Regardless of whether the drum's at two degrees or 200, the Hottop will heat or cool it until it's at 75 degrees C. This means the beans always go into a drum that's the same temperature, which should lead to more consistent roasts, something I'm prepared to wait a bit for.
The biggest drawback of the Hottop D is the lack of programmability. The stock
profile is OK for some beans but not for others, and the inability to adjust any part of the profile vastly limits what the user can do with the roast. The metal parts are great, except the whole unit gets very hot so if you
want to remove any parts oven mitts are a must - and considering the drum design is such that beans often get
stuck between the agitation paddles and the snake it's pretty much essential to remove
the pipingly hot metal end cover after each roast, then poke around with a
wooden spoon handle or something to try to remove these stuck beans so they
don't taint the next roast. It would only have taken
a larger or smaller gap to prevent this happening. The Hottop's chaff handling makes you appreciate how good a job the Gene does in this regard, which leads to another gripe - the ill-fitting of various components.
There are no gaskets on any of the drum components, nor around the exit chute
and it's quite common to have chaff sitting on the main circuit board - yes,
that $AUD200 piece of electronics. It's really poor form in my opinion and
wouldn't have taken much work to make sure the roasting area was properly sealed
off from the control area, and that some kind of gaskets were used around the
drum to keep the smoke in place. For Australian buyers, there doesn't appear to be a simple way to purchase one locally; it has to be done through their Taiwan office which means paying into an overseas bank account.July 2008: Hottop is now distributed and supported in Australia through Things Coffee.
Conclusion
So what do we take from this? Which is the best roaster? Is there in fact 'a' best roaster? Looking at the main units of this review, it's obvious from the taste tests that there isn't a clear winner. Both the Gene and the Hottop D roasters can give you bright, acidic roasts, and, in degrees, both can give you smooth, full-bodied roasts; it just depends how you use them. Generally speaking the Hottop's were more full-bodied than the Gene's and in cupping the Hottop just scraped ahead, but there was little pattern to the actual nature of the flavours and it's quite possible a different result would have been experienced with different beans.
I'm always reluctant to call a winner unless it's clear one really doesn't belong in the pack. Taking all the roasters mentioned, each has a place but some have a larger place than others:
- If I was starting out with home roasting on a budget, I'd make a Corretto-style roaster from a heat gun and a breadmaker. It may be an ugly contraption but it's capable of results equivalent to some of the Gene and Hottop roasts at a fraction of the cost.
- If you want a proper
roaster rather than a cobbled together contraption but are still price-sensitive,
the Gene is probably the pick - noting its limitations, it does the job very
well and does it at a very competitive price,
and has the bonus of Australian support (if you live in Australia!)Edit July 08: Hottop now has Australian distribution through Things Coffee. - The Hottop D probably occupies the slimmest of places in this group; if I was prepared to sacrifice the programmability of the Gene for the sake of knowing what the roast is doing and when to stop it then I'd consider the Hottop D, but really only if I was inclined to make a few home modifications (as I am) because in its stock form it's a great concept that's crippled by an inflexible control system.
- Fortunately Hottops are all designed to be upgradeable, so if I wanted the best of both worlds and money was less of an issue then I'd be definitely looking at the Hottop B or P - both do at least what the Gene does or more without many of its drawbacks - the main downside is price!
Having now used some of the new Hottop controllers and compared several new Hottop roasts to Gene roasts, the results are clearly falling in favour of the Hottop and I'm almost certain to make one of the new Hottops my weapon of choice. I'm confident that, fitted with either the B or P controller, the Hottop is capable of producing a better result than the Gene, if for not other reason than it gives you better access to the roaster's right arm - the sound of cracking beans. There are several Hottop-isms I could well live without but most of those relate to extra time taken in cleaning or waiting between roasts, and for me these are things I can live with for the sake of getting the roast I want and it will be interesting to have a closer look at these controllers in the months ahead. Others with different priorities may feel differently and prefer a different solution.
In terms of value for money, it's important to note that while the Hottop may deliver a better result, the only ones that consistently beat the Gene are also 40-60% more expensive. Considering what it can do, the Gene makes an unbeatable case of value for money, and if I only had up to $800 to spend then the Gene would be in my shopping cart.
Either roaster will do an excellent job at roasting coffee; which one is best for you depends on your priorities and your budget. Below is a brief distillation of this review which may assist you in deciding.
At a glance list of pros and cons
Corretto - typical cost $AUD100
| Pros | Cons |
|---|---|
Cheap |
Doesn't look very good |
Stereotypical buyer: A function over form home roaster who wants good results without a big outlay
Gene Cafe - RRP $AUD780
| Pros | Cons |
|---|---|
Quick back to back roasts |
Very noisy |
Stereotypical buyer: A keen small-volume roaster who's price-sensitive, who's prepared to train their ear and do their own bean cooling, or for whom three roasts in an hour is a necessity
Hottop D - RRP $AUD995
| Pros | Cons |
|---|---|
Looks more business-like than Gene |
Not programmable |
Stereotypical buyer: Someone who doesn't mind paying a bit more for a completely integrated roasting solution where they're able to clearly hear the bean cracks, who doesn't mind if it takes a bit longer from roast to roast and if there's only one temperature setting. Also suitable for someone who's prepared to make their own control system.
Hottop B / P - RRP $AUD1249 / AUD1449
| Pros | Cons |
|---|---|
Fully programmable |
(As per D model) |
Stereotypical buyer: Someone less cost-sensitive who wants a completely integrated and programmable roasting solution where they're able to clearly hear the bean cracks and who doesn't mind if it takes a bit longer from roast to roast.
* Thanks to Hottop Taiwan for assisting with the procurement of the newer roasters.
Note: As a result of our positive experience with the Hottop during this review process, Pullman Espresso Accessories is now selling Hottop and Gene roasters through its partner store, Things Coffee. As such we stand to benefit financially through the sale of any of these units through our store, however the review was written before this decision had been considered and remains as such (except any edits which are clearly marked and dated), and therefore will remain an impartial review of these roasters.
