vialroom

#lab-requests 2025-08-20

Wednesday22 messages8 participantstimes are UTC
Highlights from this day
  • tail_factor — send a capped syringe with enough for a rerun. the minimum is never enough if they need to repeat it, n of 1 obviously 20:00
  • VialBot — Batch lookup SG-1177: 5 independent reports on file, earliest 2024-10-02. 20:06
  • bea_reconstitutes — the split works from about 2 people. below that everyone pays more than the test is worth, for what its worth 20:20

sorry to jump in we publish the raw report. summaries hide the thing you needed to see

AN

i keep half the sample back so there is something to send if the first one goes missing
do not send your only vial. that has happened twice and both times it hurt

is there a queue position i can check
quantity is the test worth paying for. purity is usually fine and fill usually is not

we do not help with wording beyond that and the mods are strict about it, i think

PS

powder if you can. reconstituted travels worse and adds a variable we cannot control, anyway

[edited]
⚠️1🙏1

send powder

👀5🙏1

two labs on the same lot is a luxury but it is how we learned what inter-lab variance looks like

VB

Archive lookup: peak_split first appears in this channel on 2025-05-01.

PS

declare it as an analytical reference sample with a low value. that is what it is

TF

send a capped syringe with enough for a rerun. the minimum is never enough if they need to repeat it, n of 1 obviously

👀7😂1
VB

Batch lookup SG-1177: 5 independent reports on file, earliest 2024-10-02.

🎉10
PS

i coordinate the round, collect the codes, and post every result including the dull ones

DD

a pooled result tells you about the pool. it does not tell you about your vial

BR

the split works from about 2 people. below that everyone pays more than the test is worth, for what its worth

🎉8📈16👀1
GG

i would rather fund one careful quantity test than three purity tests

📉5

label the syringe

HH

the queue is posted and updated. if your code is not on it, it did not arrive

HH

submissions go out blinded. the lab gets a code, not a supplier name. that removes one variable, it is not distrust