Candy 25 cabinets — hell on Earth, part 2

In late 2013 I wrote a blog post describing the negatives of owning one of SNK’s Candy 25 cabinets. I wanted to follow up to that with an incident that happened today, adding further justification.

For the first time in almost 10 years I had an interested buyer in the cab. A mutual friend introduced us, and both of them visited. The potential buyer seemed incredibly interested. Said mutual friend wanted to try out (or demonstrate) the fact that the cabinet had a rotatable monitor, and proceeded to do exactly that. This is something I had done before (maybe back in 2001) without issue, but I was very very careful about it, but since then had always run my display horizontally, not vertically. So I hadn’t rotated it for a very long time.

What happened today: a mounting bolt on the rotating chassis (which holds the CRT) didn’t have full clearance, and ended up smacking into the Toei TC-RM25T CRT chassis (this is what contains all electronics that drives the CRT), specifically the large transformer (not the flyback) — and it bent/tilted quite severely. It was with enough force that something between the transformer and PCB became damaged; that is to say, the force used likely broke or damaged one of the pins that kept the transformer on the PCB. The clearance was short by about 1/16th of an inch.

Upon powering on, the CRT or its electrical circuit emits a very high pitch yet slightly variable-frequency noise, and nothing happens past that point.

Visual inspection showed no actual damage to the transformer or the pins, but very close examination showed that the PCB itself may have cracked as well (2 traces showed signs of possible cracking). We verified continuity with a multimeter that those traces were OK (but that’s just continuity — the damage may be enough that voltage flowing across those traces is no longer reliable). We also tried re-flowing a couple of the solder points on the PCB relating to those traces. No improvement.

Accidents happen, and I consider this exactly that. But how could it have been avoided? Surely SNK wouldn’t make a cabinet prone to this type of problem, would they? No, they wouldn’t. How was that bolt too long for clearance? I hadn’t changed a single thing about the CRT or related parts since I bought it. We’ve only been able to reach two conclusions:

1. The previous owner of the cabinet was an arcade or gaming operator himself. As we began to inspect the CRT chassis itself, we found that the bolt had been put on backwards — that is to say, the long end of the bolt was facing towards the TC-RM25T, rather than the other way around. In fact, several bolts and screws were found to be like this. Additionally, several screws looked to be non-standard (as in the previous owner had likely lost screws and found makeshift screws/bolts/whatever that fit the threading but were never cut down to proper lengths).

2. When I bought the cabinet, it was shipped to me via freight from southern California. It arrived at my workplace, which is where it sat (and got used). This was also the place where we tested the monitor rotation capability (without issue or clearance problems). Since then, the cabinet has been moved twice: once to my apartment, and later from there to the garage of my current abode. It’s very likely that the cabinet (especially during the 2nd move, which is how and when I threw my back out) had been jostled severely and shifted around, maybe just enough where clearance became a problem. Since I’d never rotated the monitor since that one time, it wasn’t something I ever discovered or came across.

So I now have a non-working Candy 25 cabinet due to a somehow-badly-damaged Toei TC-RM25T CRT chassis.

Finding a replacement CRT chassis is, needless to say, a serious pain in the fucking ass. First of all, nobody I can find sells the TC-RM25T. Instead, the only one you can find is the (older) TC-RM251S. Next, is it fully compatible with my tube? I don’t know. Does it need or use a different flyback than the TC-RM25T? I don’t know. Next, how much does a TC-RM251S cost? The guy who runs arcadecup.com sells them for US$150 (or US$175 if you need the flyback), plus shipping. And then: how easy it is to replace/install? It’s not an issue of easy vs. hard, it’s an issue of safety: the CRT chassis directly connects to the CRT tube via an anode and anode cup, which is literally the most dangerous aspect of a CRT (runner up is high-voltage circuitry and capacitors that hold charge for months even after being powered off). There is no way in hell I would attempt to disconnect the anode cup because of the danger — I could literally die. Considering I’ve had bad luck nearly my entire life, yeah, NOT doing this myself is the obvious and safe choice. Leave it to a professional. These are just a few of the complications.

In other words: replacing the CRT chassis would require me to spend probably US$175 for the chassis, then pay someone for the labour (which is expensive, especially considering I live in Silicon Valley). If I chose to have the TC-RM25T repaired (vs. replaced), I’d have to pay for that labour *twice* — once for removal (of the damaged chassis), and once for installation (post-repair).

So what am I going to do? I don’t know. Financially it doesn’t make any sense to have anything repaired or replaced. For about as much as I could sell the cabinet for (I was hoping for US$300), it’d cost me to have the parts repaired or replaced and labour (maybe even a bit more than that). In short, I don’t know what I’m going to do. To be honest, it’s actually more cost-savvy to just pay someone to pick the thing up and haul it to a landfill and pay the enormous dumping fee. Seriously.

This is just another example of why these things are a fucking money sink. They’re only plausible (and economical) to own if you have the capability to get parts *and* know how to repair them yourself. Fuck arcade cabinets. Never again.

Moto G 2015 (3rd Gen) vs. 2013 (1st Gen)

Since mid-2014 I’ve owned a mobile phone: a Motorola Moto G 2013 (1st Generation model). When shopping for a phone (since the last time I owned a mobile was in 2002 or thereabouts), the requirements were pretty simple: low-cost (I cannot justify spending US$600), worked with T-Mobile (i.e. used GSM), and was something I could buy immediately (vs. buying it through the carrier + paying monthly charges). My friend Liz recommended the Moto G.

In May 2014 I purchased the 16GB GSM model from Amazon for US$199.95, and I’ve been mostly happy with it (my biggest complaint has been the selfie camera brightness bug introduced in KitKat 4.4.3, which still exists even as of Lollipop 5.1). I was also a Motorola beta tester of Lollipop (I had a lot of UI-related complaints, some of which got rectified, others didn’t).

However, odds are the Moto G 2013 won’t get Marshmallow 6.0 (another example of Google and vendors focused excessively on their cash cow of forcing people to buy new phones just to get software updates), and I had begun to encounter several applications which really didn’t perform very well. And that selfie camera bug really pissed me off (yup, I’m OCD about bugs).

I began looking for a replacement phone with the same requirements as before, and of course was introduced to the Moto G 2015 (3rd Generation model). The 8GB model was US$179.99, while the 16GB model was US$219.99. The 2015 model included a microSD slot (the 2013 model doesn’t have one), and after doing some research I found that with Marshmallow you’d be able to use your microSD card for system space/storage; it didn’t make much sense to me to spend US$40 to get 8GB of space when I could spend US$14 and get a high-end 32GB microSD card. After reading several reviews of the 2015 model, I ended up purchasing the 8GB version from Amazon. The first one I received actually had dead pixels and a touch screen that seemed non-responsive about 20% of the time (more on that in a moment), so I got a replacement (which had no dead pixels).

Below are the reasons why I’ve chosen to return the 2015 model and stick with my 2013. This is literally the only “review” of this sort — specifically giving a negative opinion rather than a positive — that I could find, so apparently I’m in the minority.

Continue reading

Candy 25 cabinets — hell on Earth

I’m the not-so-proud owner of an SNK Candy 25 arcade cabinet. Why not-so-proud? I’ll get to that. But first, some history and education.

Foremost: I am not talking about the Neo Candy 25 cabinet. Despite the similar name and similar look, they’re actually quite different (the Neo is a lot easier to work on, especially if having to work on anything relating to the coin mechanisms).

Continue reading

Instructions for building TomatoUSB Toastman RT-N firmwares

Back in September/October I had the “pleasure” of dealing with trying to build TomatoUSB (specifically Toastman’s RT-N releases) on a Linux system I have (thank god for VMware Workstation!). I was given all sorts of reference materials from folks on the linksysinfo.org forum, except I kept running into all sorts of problems. I tried other Linux distributions, other releases of the same distribution, etc. and the failures all seemed to differ.

I figured it would be worthwhile to document exactly how I got the firmware to build and what the necessary steps are as of the date of this blog post.

Continue reading

Hard disk drive heights and polypropylene storage containers

Given how many hard disks I tend to have, I recently decided to purchase some anti-static polypropylene foam containers to store my drives.

Common retail outlets (Amazon and others) were selling these containers (which would hold between 10 and 24 disks) but for crazy prices: US$100, US$130, and US$150. And some other retail vendors want US$425.

The price was simply too steep; I wasn’t needing something brand new, used would be fine. So I turned to the one place where stuff like this tends to end up being sold for reasonable prices: eBay.

Continue reading

Examining the WD10EFRX

Today I received a WD10EFRX, otherwise known as the 1TB model of Western Digital’s new Caviar Red drives…

The colour red, for me, is associated with two things: stop/halt signs and blood — two things which should definitely not be associated with a mechanical hard disk. :-) I say that in jest, but still…

The new Red drives boast “full NAS compatibility” or “NASware”. This is just more marketing buzzword-speak for something that does not need such segregation. A hard disk should work in any system and not induce problems or inhibit I/O) while providing features that the WD Caviar Green drives have (lower temperatures, “IntelliPark”, etc.), but still providing good throughput (not as great as the WD Caviar Black, but supposedly better than the Green). It’s very important to remember that companies like Western Digital created this delineation in the first place, so it’s very ironic (and depressing) that they’ve now begun to market models of hard disks that “solve these problems” when they themselves created the problem to begin with. Phrased differently: MHDDs used to work in anything you put them in (desktops, servers, etc.), with zero repercussions or issues, and it was the MHDD manufacturers themselves who created these firmware-level “features” (issues). It’s almost like the drives of today are being advertised/marketed with the assumption that those of us who were alive 13 years ago would hopefully have forgotten about how hard disks used to Just Work(tm). Well, I’m one of many who hasn’t forgotten.

Anyway, I’ve wanted to get my hands on one of these drives since they were announced, because I wanted to see feature-wise and behaviour-wise if these drives behaved better than the Green drives (you can read my “review” of present-day Green drives and how firmware-level features in them that do nothing other than frustrate and cause major problems).

Specifically, I wanted to examine the following things:

  1. Platter count and reliability
  2. What ATA-level and firmware features the drive supported, and which features were enabled (or disabled) by default
  3. If the infamous Load Cycle Count (LCC) issue existed on these drives
  4. Very quick performance testing (I am not OCD over benchmarks)

So let’s get to it!

Continue reading