Brother: Updates aren’t behind degradation of quality or removal of features.

  • IHeartBadCode@fedia.io
    link
    fedilink
    arrow-up
    12
    ·
    12 hours ago

    This one is a difficult 3rd party toner isn’t all formulated the same. The main carrier, the thing that picks up the charge and moves the toner onto the paper is usually a blend of polyester and/or polypropylene. Various mixtures have different properties in the presence of a static electric charge.

    The charge on the drum and exposure time to the laser are all parameters that are tuned in the firmware and it’s likely Brother would tune those parameters to toner they’ve formulated. So getting sub performance from 3rd party toner can be a hit or miss, because some will have a batch of toner that hits the mark and then the exact same group could have a batch that pulls poorly on the toner. A lot of 3rd party toner is seat of your pants kind of operations with minimum QA process. But most people using 3rd party toner is fine with “good enough” quality.

    So even using the same maker of toner isn’t some golden seal of accuracy that you’ll get the same formulation every time. Now bricking functions is concerning and that’s something to keep an eye out for. I’ve got three different printers in the office, all Brother, all using 3rd party, none having bricking issues. But that’s completely anecdotal. Likewise, I’ve only ever heard maybe one or two claims online by folks of bricking Brother printers.

    Now don’t get me wrong, I really enjoy Louis Rossman’s videos. But that he’s going off completely wiki data is suspect. Now as for 20Factorial, yes, that is Brother’s standard line for bad color registration to ask for you to swap over to OEM toner. So for those not in the know print registration … well it helps to understand in a color laser printer, you usually have four toner tanks (Yellow, Cyan, Magenta, and Black). Now all those tanks are located in physically different places inside the printer and so there needs to be a way for the printer to understand those physical differences and ensure that the colors overlap each other perfectly to get the correct mix of colors to print what’s being asked. That’s what print registration is, sometimes that’s called color print alignment.

    Brother’s MFC-3750 has an “automatic” process by which this is done where a chip passes along information to the printer on the properties of the mixture of toner within the tank. This is where you can run into an issue because the automatic information might be just a copy the 3rd party picked off a used tank and the data passed to the printer matches nothing in terms of the formulation the 3rd party put inside the tank. And a newer firmware might see that copied info as out of spec formulation. Who knows? That’s the gamble with 3rd party you take.

    Brother and any other printer maker gets complaints whenever their machines don’t print correctly and sometimes that can be the formulation of the toner. Sometimes it’s a firmware update that tightens the requirements for reported formulations. Sometimes that can be malicious and sometimes that’s just a side effect of tweaking how the firmware processes different formulas. It’s a lot of variables.

    So that part about 20Factorial’s printer is something that “sucks when it happens” but not all toner is the same thing. Brother can only assure proper assistance with mixtures they’ve had in their labs. BUT, and a big but here, while not all toner is the same, there’s tolerance here and usually 90% of all toner on this planet falls within that tolerance. But you can never tell with 3rd party toner, it could be a one off batch, or yeah it could have been a firmware update that lost it’s damn mind when it got presented with the chip information from the 3rd party toner.

    The whole point is, this is really hard to nail down.

    Now what isn’t is when shit starts bricking. That’s 99% of the time the OEM doing that shit to you. Now that could be on purpose or sometimes it’s bad error checking for 3rd party stuff, but it’s absolutely the code from the OEM doing it to you. I will say automatic color registration is a tricky bitch to get right, even when using OEM toner.

    I’m not going to defend Brother here if they’re doing shady stuff. But all the claims except fncamo, have that feel of losing the gamble with 3rd party toner, 99% of the time you win, but there’s always that 1% that gets you.

    fncamo’s claim of bricking is odd, but at the same time, they may have power cycled while in a critical point of the inline firmware programming and bricked their printer themselves. But maybe not, it absolutely is the only one out of all of this that got my attention. I wish there was more to their comment.

    But 3rd party toner SHOULD work, but only OEM toner WILL work. Like I said, 99% of the time the toner is chemically fine. There could be something else to what happened for the others.

  • earphone843
    link
    fedilink
    English
    arrow-up
    12
    ·
    14 hours ago

    So their proof is like 3 reddit comments spanning 3 years?

  • troed@fedia.io
    link
    fedilink
    arrow-up
    4
    arrow-down
    2
    ·
    11 hours ago

    Yeah twice now in short succession people who really should know better have caused huge dramas around nothing burgers.

    Firefox isn’t selling your data

    Brother are still the only printers to buy