General

References? E.g. CMS, LHC, numbers, …?
Is there a reason not to write Figure but always use abbreviation? It’s consistent but seems unusual.
Table of contents necessary in this kind of paper?
Lots of jargon used, explain CMS internal phrases
Sentences seem to contradict each other sometimes (cf. e.g. l. 110 type B comment)
Consistency whether something is written with/without dash, capital/small
Introduce naming before describing things


Type A (English/Style/Format/Figures)

In TOC: Why muon/Muon & trigger/Trigger sometimes capital, sometimes not?
l. 26: its experiments
ll. 27/193/194/197: “about/around” instead of approx sign
l. 81/121: in l. 37, “in [1]” and “in [2]” are used, here “in Ref. [n]” → should be consistent
l. 195: the remaining 51 unprescaled algorithms
Tab. 1: Caption: everywhere else, it is “Level-1 trigger”, here, the dash is missing. Last line in the table itself: Should be “Sections 7.2 and 7.3”
Tab. 2: In comparison to tab. 1: “most used” or “most-used”? Also “Level-1” or “Level 1”?
Start Section 5: Inconvenient to read two lines before four pages of figures and tables → maybe put it on page 10 to start after figure 4?

Type B (Content/Structure/Strategy)
Abstract: Why start with 2015? How is this linked to trigger requirements?
l. 38: the LHC energy was increased (same for lumi: no decision by the machine, not yet at least...)
l. 58: no need for double reference
l. 66: the magnetic field does not really track the particle, it bends the trajectories
ll. 77-81: seems slightly redundant; more or less exactly like in section 1; last sentence should be in a new paragraph anyhow
l. 95: numbers can not all be correct: 2556/48=53.25… There are empty bunch places, yes, but no non-colliding bunches, are there? (probably type B but perhaps I just misunderstood the sentence, then it should be type A) Also in Fig. 3, it quotes a fill with 2544 bunches which would fit
e.g. l. 105: isn’t an event one pp interaction? I.e. in one bunch crossing, there are multiple events (PU) but an event should not have PU/several interactions
l. 110: in ll. 105/111 and Fig 3, it could be understood that the inst. lumi limits the PU but here it says the contrary → what is correct or how is the relation?
l. 115: trigger menu not introduced before
l. 143: pT not introduced before
Section 3: The last paragraph is obviously making a lot of sense in the context of this paper but the rest of this section seems only loosely linked → is it needed? And if, how does it connect to the trigger performance?
l. 167: why “unbiased” (so why “”)? Isn’t it unbiased?
l. 206: perhaps also paragraph before: quote these standards/technologies – they are not explained anywhere (or leave it out completely since it only refers to Run 1)
l. 217: is trigger tower explained anywhere?
l. 218: Why same value in eta and phi? Is there a reason? What does “most” mean? What about the others?
l. 219: What is a scheduling algorithm and why would we need it? And what does not complex mean?
ll. 221+: what is a complex algorithm then? And what is the minimum amount of data?
l. 234: what is a Mezzanine Card? What does it do? And how is the AMC13 different from the trigger AMC to which it provides signals (l. 235)?
Fig. 1: “multi” → “double” or are more than 2 objects allowed?
Fig. 3: Caption sentences seem not to match each other: A rate is shown, not a prescale value. The link to the instantaneous luminosity is unclear. If quadratic functions fit well, why point out that the dependency is mostly linear?