New Convoy C8 – Clearly better

Nice pics TK! I did embignify and indeed found great justice. :smiley:

To “clear” up a few things for people…there is no silver dye.

Clear anodizing is literally the absence of any added dye.

What people are seeing as “silver” or a glittery/metalic surface is just light reflection. These hosts are media blasted after the initial CNC machining. This is done to smooth out the surface and soften or hide the machining marks. The media blasting process creates the blasted surface (TV snow) you are seeing and the light reflects off these tiny little craters in the metal. The black anodizing is thicker (HAIII) so you see little or none of these little marks in the finished product. On the clear as well as the lighter colors (red, blue, green) you will also see a bit of this “metallic effect” as those colors are (HAII) as the third coating to bring it to HAIII makes the color very dark.

Simon is doing a run of clear anodized S2+ now that are not going to be media blasted after machining. Those will look much more like the bare aluminum A6 you showed us. :+1: Many people (myself included) have asked for this. Some of us want to see those machining lines. When a light is well machined some of us want to see that craftsmanship not on every light of course. But it’s a nice new look to add to our collections. Depending on how well it turns out more clear ano lights may be skipping the media blasting process in favor of this more “raw” look.

Cheers!

I’ve never hard any reports of this before yours but I will pass this along to Simon. If there is every any concern of quality in any Convoy product please don’t hesitate to contact Simon directly through the Aliexpress store. Simon has expressed to me many times that he wants people to contact him directly any time there is an issue. He will always do his best to make things right.

I actually rather like the bead blast process look, but omitting that step won’t stop me from buying a clear anno light.

As far as the tube surface conductivity goes, I don’t see it as a big issue personally. I just assume all of my tubes are conductive and use my batteries accordingly.

If you have a protected cell with a damaged wrapper DO NOT USE IT regardless of the annodization. The tabs coming from the positive end tend to have sharp edges, and I would think they could scratch through the anno fairly easily if they can rattle inside the tube at all.

Also, don’t use ANY damaged wrapper cells in 2S lights, protected or not. If the top cell closest to the head has a damaged wrapper and touches the inside of the tube it will effectively short the bottom cell once the switch is pressed.

PVC heat shrink is cheap, if you have a damaged wrapper, cut it off and re-wrap with a new one. A little protection now can prevent a potential disaster later.

In a Titanium light or a full copper light, as well as Brass or Stainless Steel lights, there is always continuity. Not a deal breaker just a twist in the game that must be acknowledged and worked with.

A great review from mhanlen : (Sub $25) Convoy C8 Clear Flashlight review Best Budget Thrower Around Now Updated! - YouTube

That might explain both the color and the spots with continuity. If the anodized layer is really thin, it could still appear silver… and might have areas thin enough to conduct. Generally anodizing darkens the surface, so I figured it would need a silver dye to lighten things up again.

I’ll have to check with Simon on the details but I think you may be spot-on. I remember him being disappointed in the first several attempts at clear anodizing because it was too dark. He said it looked almost like the grey color he was already selling. It could be that they had to go with a thinner anodization layer to achieve the desired look.

My black 3/5 mode C8 has a feature which I find annoying: when the light has been in a particular mode for more than a few seconds, doing a half-press does not cycle it to the next mode. It just blinks off and comes back in the same mode. I guess it is supposed to be a ‘mute’ function but I don’t care for it.

Does the new C8 with biscotti have this same quirk ?

All Convoy lights with that driver respond that way. The intent is to guard against unintended half-presses changing the mode after the user has selected a mode they intend to use for a while.

If you half-press twice in rapid succession, the light will change to the next mode, with subsequent single half-presses also changing mode until the light is left alone for a few seconds again.

I don’t think so, but I don’t have one of those lights, so you might be best waiting for someone else to confirm that.

Yes, I guessed there was some intention behind it (rather than a bug). I don’t like it personally, I prefer a flashlight that always does the same thing when a button is pushed.

I am thinking of buying the clear C8 so hopefully it doesn’t have it.

The new ones with Biscotti don’t do that. When we started using he Off-time capacitor it was to do away with that. All the old Nanjg 105’s and Qlites had that type of action, you had to bump it twice to nudge it forward after it’d been on for a bit.

Biscotti in the new Convoy’s brings a wealth of change, virtually all good. :wink:

As far as I can tell, there wasn’t actually any intent behind that. It was an unfortunate side effect of the way it implemented memory.

The “on-time memory” it used worked like this:

  1. Light turns on. Read the mode number from eeprom, and go to it.
  2. Write the mode number plus one to eeprom.
  3. Wait a second or two.
  4. Write the current mode number to eeprom.
  5. Wait forever.

So, if you leave it on for more than a second or two, next time you use it it’ll go to the same mode. If it was on for less than a second, it goes to the next mode. It doesn’t attempt to measure how long it was off.

Biscotti uses off-time memory instead. The behavior is determined by how long the light was off. So, during use, a quick tap always goes to the next mode. This is almost universally preferred. The main downside is that it’s more sensitive to what happens while the power is off, so a lighted tailcap can interfere with the “off time” measurement.

Very interesting, thanks for explanation DB Custom and Toykeeper.

If it is writing something to eeprom every time we switch modes, could the memory eventually ‘wear out’ ? Or is eeprom write endurance sufficiently high for this not to be a problem during a typical lifespan ?

I don’t know if the old 3/5-mode firmware uses wear leveling. If so, the ROM should take at least a few centuries to wear out. If not, it should still take at least a decade. And those are both low estimates. Biscotti should take nearly a thousand years, even with quite a bit of daily use, before it ever starts thinking about having eeprom wear issues.

<—— filling out post-it note that TK guarantee’s Biscotti for 1000 years, logging it at 4/28/2017.

I’ll be watching you! :stuck_out_tongue:

Dale, I’m afraid to see what your post count will be 1000 years from now. I’m guessing it’ll cause an overflow at some point :stuck_out_tongue:

?huh ,tnuoc ym fo emos ecuder ot redro ni sdrawkcab gnihtyreve gniyas trats ylbaborp dluohs I

I don’t think it works that way :partying_face:

(man, after a bit of Sobieski, that took a minute to process!)

ROFLMAO!

I had the standard text on standby, just in case. :stuck_out_tongue:

Y’all are soon to be on vacation from my ever present wit so enjoy it while you can. :wink:

So, the eeprom is rated for 100,000 erase/write cycles minimum per byte… but in actual testing people tend not to encounter any failures until more like 10 million write cycles. So take 100k as a lower bound and 10M as an upper bound.

With only 32 bytes used for wear-leveling, let’s find out how many times a day you’d have to press the button to make it fail in 1000 years or less:

  • 100,000 * 32 / 365.24 / 1000 = 8.76
  • 10,000,000 * 32 / 365.24 / 1000 = 876

So… to make it fail in 1000 years, you’d have to press the button at least 8.76 times per day (every day) or potentially up to 876 times per day. I’m pretty sure some other part of the light would wear out first, like … the switch.