What type/kind of "analog 8-track" are you using? One of the nicer elements of using analog tape is what it can do for drums - providing the machine is a real pro recorder and not some little POS 1/4" Fostex or Tascam...
There are no set rules on compression, but a little on kick goes a long way. Some people spank the snare, it all depends. I try to leave things alone and watch my levels VERY carefully. Compression can make the room a bigger part of the equation, and drums in a less-than-great acoustical environment can sound terrible with a "little compression". Use your ears and (again) watch your levels.

I use 8 behringer compressors and I compress all drums going down. Each to his own.

I've made plenty of recordings on 1 inch analog recorders back in the day. I always added some bass drum compression. I always added some snare drum compression. I would also record that rather hot to also have the tape saturated. I recorded tom-toms & overheads, frequently, without any dynamics compression. I would also lay all of the tracks of the drums on to separate tracks with the overheads and tom-toms being combined. Upon finishing the initial tracking, it would then be necessary to bounce down to just 1 or 2 tracks. In that process, I would then add some fast downward expansion a.k.a. noise gating on the bass drum & snare drum. On occasion, I would even go as far as doing that to isolated tom-tom tracks as well. In the end, your drums are now down 2 analog generations and in and out of the console twice. This process generally require that one would utilize the record/sync head to playback through. The only caveat is that you cannot bounce to an adjacent track. Well, you could but you could not do it at full level. So a guard band track would be necessary. The only way around that to go to an adjacent track was to utilize the playback head. But then you would also have to do the same with the other recorded tracks to maintain synchronization. Although it's all doable in one form or another. And even upon those bounces, you may still need to be adding some short or gated reverb to snare drums and such. So you're continually mixing during the entire recording process. Noise gates/downward expanders were also important in trying to maintain decent signal to noise ratios as I never liked what Dolby nor DBX sounded like even though they worked. Yes, I did on occasion utilize Dolby A & DBX type I and even DBX type II, on TA-SCAM recorders since they had vastly narrower track widths and you could also never run them at 30 IPS. So a lot of this really depends upon the recorder in use. DBX makes the recorder's frequency response twice as bad as without it. And on a TA-SCAM recorder it always really was ± 3DB 50-15,000 before noise reduction & ± 6 DB after noise reduction. That's a huge bad sounding variable. Sometimes though, the peaks and valleys would fall into areas you'd want to EQ that way. So it was a crapshoot with your final results.

So I don't necessarily think it's prudent utilizing all 8 of your compressors during tracking especially since you cannot stereos strap all of them together as a single 8 channel compressor. This will get you into lots of acoustical trouble. But if you were to also include 8 gates, you may find it much better sounding when drums are only on for a split second while played. But Gates on overheads generally don't work at all effectively or in a pleasant audible manner. Unless you are utilizing gating for just the room ambient pickup of the drums just to thwart the hanging acoustical unpleasantries while making the drums sound huge.

And frequently, I would include plenty of boost equalization on drum tracks while recording since doing it on playback only increased the noise. Cut equalization can be beneficial on playback/bouncing to also control noise and to tailor the drums more effectively. I'd be happy to post a song done exactly like this from 1979 if you'd like to hear it? It was, however, recorded upon a MCI JH 110A-8 at 30 IPS, barefoot to Scotch 250 at +6 DB which is actually +9 DB over standard 185 nano webers per meter reference level which is actually, +6 DB over the 250 nano weber per meter reference level of most later calibration tapes. People get quite confused over calibrating their machines when they think a 185 nano weber per meter calibration tape at 0 VU is lower than a " elevated level " calibration tape that has a 250 nano weber per meter 0 VU reference standard. It's actually backwards. 250 nano weber per meter 0 VU is actually -3 DB lower in level than 185 nano weber per meter reference. In that way, by lowering playback level, you are increasing record level in comparison and hitting the tape harder. And a 320 nano weber per meter 0 VU reference level was actually -6 DB lower than 185 nano webers per meter.

Then there was also bias level to deal with. Depending upon the tape formulation & construction of the record head, YOU had to decide how you wanted your tape to sound, perform & respond. The lower the bias level the less it would cause self erasure, extended high-frequency response, sound more open. But you couldn't hit the tape as hard as the saturation would not sound as kind. So you would slightly over bias. This would allow for better sounding saturation while losing a little bit of high-end response above 15 kHz. And when recording bass guitar, if the bias wasn't quite right, you would hear what would sound like gravel being walked on in the background of the bass guitar. A.k.a. " Bias Rocks ". So sometimes, you would want to tweak bias while listening to a bass guitar while recording. You'd go for the least amount of gravel. That form of biasing was also popularized by John Stephens while trying to record a 10 Hz sine wave. Yup, I said 10 Hz and no recorder can record 10 Hz. It ain't possible. You hit the tape hard with 10 Hz and while only listening through headphones (not speakers) you tweak the bias for least modulation noise or rocks. Which generally always equated to +2 or, +3 DB over bias at 10 kHz, respectively. Most low noise & high output tapes generally specified +2 DB over at 10 kHz while the hotter tapes yet, required +3 DB over at 10 kHz. And those settings generally equate to -.5 DB at 1 kHz. All of these suggested adjustments were generally performed at 15 IPS and/or 30 IPS. It's different when you're dealing with 7.5 IPS and lower. Because of the increase in high-frequency preemphasis during recording of the record electronics.

These are just general suggestions from my experiences with Scotch/Ampex & Agfa recording tape of the 206-207/406-407 & 226/456 of 3M & Ampex formulations. 3M/Ampex/Agfa also had 250/499/468 series of formulations which required the higher bias and could accept signals +3 DB hotter than the previous ones. There were later formulations that can even accept hotter levels than those. There is some confusion as to the tapes currently available and their basic similar formulations to those previous now discontinued manufacturers. Not all of those hotter tapes were fully compatible with older machines. A loss of depth of the erasure was usually the result and no amount of bias increase could compensate for that. So on some older machines like Scully's, you would be restricted to formulations more similar to 3M 206/226, Ampex 406/456. Newer machines by Ampex & 3M also Studer had no problems with the latest formulations of tapes. For further information feel free to PM me about that.

When/if talking about microphones for drum recording, SM57 & Sennheiser 421 were the staple drum microphones. While, numerous condenser types were usually utilized for overheads. This was not a 100% de facto standard however. I knew plenty of engineers that utilized Neumann U-47 FET's on bass drum. I knew plenty that utilized Sony & AKG Condensers on tom-toms. SHURE SM 81's & AKG451's on the bottom and even topside of snare drums provided the microphone pad switch was enabled or the pad was screwed on to the 451's. Otherwise it's nasty overload time. And you still might have needed the microphone preamp pad engaged as well.

So that's a lot of technical blah blah about recording drums to analog tape. You don't have to follow all of that as some of that is beyond the scope of many home recording enthusiasts. But if you are aware of what we actually had to do back in the day, they'll be less surprises and poorer outcomes for you. Tape saturation on drums can be a many splendid thing but it can also be overdone leading to less than desirable results. Essentially, our VU meters were regularly banging into the red with every hit of those drums on their respective track to obtain the natural limiting characteristics of the tape saturation. That's why utilizing limiters/compressors with too fast an attack time can be completely counterproductive in this application. But when you want that ultra-fat bass drum and totally snappy snare drum, some dynamic range limiting along with the saturation gets you that sound. You just have to learn the art of how much of each to best utilize. That only comes from lots of experimentation and listening. Yet some of this is also possible when utilizing even drum machines that already have superior drum samples complete with processing already. It's all in how you want to cook it, rare, medium rare, medium, medium well & well done. You most likely won't like the sound of burnt drums?

SMACK! You're it! Or, rather, you're there.
Mx. Remy Ann David

I just finished tracking drums to 1", don't hit them as hard as you would think, I try to stay around zero, for me it gets the best natural sound

really?

All that to record drums to analog. I did thousands (probably) of 24-trk sessions, and I would never go into the red with any drum because VUs don't track transients, stay around -10 dB on a VU, or barely touch O with an LED meter.

You do not need to compress drums as they go down, and you can record kick, snare, and combine Loh Roh with the toms on trks 3 & 4 and then bounce them all to 7&8 stereo. There's no adjacent track crosstalk problem that way - or for that matter, you could go right to tracks 5 & 6 as long as track 4 (right) is going to track 6 (right) - see (it's not adjacent that way)?

Bias your tape to manufacturers settings, if you use the same kind of tape all the time you probably don't have to worry about that, but either way it is not that hard to calibrate a tape deck as long as you have recorded reference tones to setup the playback calibration first.

I would mic all the drums - the main thing is mic placement, and I could write a book about that. The trick is to keep mics pulled back a few inches - like on a snare drum you want to mic the shell from 3 to 8 inches back, so you are getting some top head and and some bottom. If you put the mic over the top-head, you hear no snares, put it on the bottom you hear no pop. See? On the rest of the drums, try to point at where the stick hits the head (but keep the mic back).

Never mic a hi-hat from the side, you get air rushing out, Mic from above at the point where stick hits, but not directly above, more like a few inches out, so you hear the hat working from the foot pedal, but not air. Never ever mic cymbals from the sides, always from above.

What else?

I'm willing to bet the OP is talking about one of these:

Cheers duh

Attached files