From: "Donor, John" Subject: Re: [sloan-25mlog 5972] 2.5m obslog 20220331 (SJD 59670) Date: Fri, 1 Apr 2022 18:18:35 +0000 --_000_238089be2bf94c0298ba576eeb0cb647tcuedu_ Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: quoted-printable Ok, got it. I think that's all clear to me! Sending to Kaike to make sure o= bservers are in the loop. Thanks, John ________________________________ From: Jonathan Trump Sent: Friday, April 1, 2022 12:52:01 PM To: Donor, John Cc: obs-report-25m@apo.nmsu.edu; Kevin Covey Subject: Re: [sloan-25mlog 5972] 2.5m obslog 20220331 (SJD 59670) [EXTERNAL EMAIL WARNING] DO NOT CLICK LINKS or open attachments unless you = recognize the sender and know the content is safe. Hi John, Small follow-up: if they had time to get another exposure on 112359 to impr= ove on b1_sn2~18.99, would you have wanted that extra exposure? I'm not sur= e if this an efficiency vs quality trade-off, or if the "90% of the goal" i= s a bit of a compromise, or..? I don't have a good sense of the trade-offs = at all, haha. But I think it's important that we know the answer to that ex= tra exposure question and a basic reason why. Yes - I still want to achieve 100% of the SNR2 goal within an epoch if poss= ible. The choice to make >90% complete should only be made if the epoch wou= ld be extended on another night. In other words I see this as a compromise = between hitting the SNR2 goal and between splitting an epoch over more than= one night. While I'm talking: I want to put on my MWM hat and ask if the 90% requireme= nt for AQMES would impact "ride-along" (potentially single-design) targets = significantly? Since that could potentially be ~half the signal on one of t= he designs, likely the last one, I suspect we can't ignore that considerati= on. I think this has been brought up a few times, just don't want to lose t= he thread. Yes, this is very similar to the issue with the BHM point of view of "epoch= completeness" superseding "design completeness" (as discussed Tuesday!). T= he AQMES-Medium fields (10x4) all seem to accomplish the single-design targ= ets pretty early in their total number of designs, so I'm pretty confident = that early epoch completion on them would not adversely affect design compl= etion of non-AQMES targets in the field. (Assuming we pay attention to inco= mplete designs in year 2+ survey planning.) AQMES-Wide is more complicated = - I should probably do the exercise of checking robostrategy to see how man= y targets would potentially be affected by early epoch completion in the 8t= h designs of 2x4 fields. best, Jon --_000_238089be2bf94c0298ba576eeb0cb647tcuedu_ Content-Type: text/html; charset="us-ascii" Content-Transfer-Encoding: quoted-printable Ok, got it. I think that's all clear to me! Sending to Kaike to make sure o= bservers are in the loop. Thanks, John From: Jonathan Trump Sent: Friday, April 1, 2022 12:52:01 PM To: Donor, John Cc: obs-report-25m@apo.nmsu.edu; Kevin Covey Subject: Re: [sloan-25mlog 5972] 2.5m obslog 20220331 (SJD 59670) [EXTERNAL EMAIL WARNI= NG] DO NOT CLICK LINKS or open attachments unless you recogni= ze the sender and know the content is safe. Hi John, Small follow-up: if they had time to get another exposure on 112359 to i= mprove on b1_sn2~18.99, would you have wanted that extra exposure? I'm not = sure if this an efficiency vs quality trade-off, or if the "90% of the= goal" is a bit of a compromise, or..? I don't have a good sense of the trade-offs at all, haha. But I think it's= important that we know the answer to that extra exposure question and a ba= sic reason why. Yes - I still want to achieve 100% of the SNR2 goal within an epoch if= possible. The choice to make >90% complete should only be made if the e= poch would be extended on another night. In other words I see this as a com= promise between hitting the SNR2 goal and between splitting an epoch over more than one night. While I'm talking: I want to put on my MW= M hat and ask if the 90% requirement for AQMES would impact "ride-alon= g" (potentially single-design) targets significantly? Since that could= potentially be ~half the signal on one of the designs, likely the last one, I suspect we can't ignore that consideration= . I think this has been brought up a few times, just don't want to lose the= thread. Yes, this is very similar to the issue with the BHM point of view of &= quot;epoch completeness" superseding "design completeness" (= as discussed Tuesday!). The AQMES-Medium fields (10x4) all seem to accompli= sh the single-design targets pretty early in their total number of designs, so I'm pretty confident that early epoch completion on = them would not adversely affect design completion of non-AQMES targets in t= he field. (Assuming we pay attention to incomplete designs in year 2+ s= urvey planning.) AQMES-Wide is more complicated - I should probably do the exercise of checking robostrategy t= o see how many targets would potentially be affected by early epoch complet= ion in the 8th designs of 2x4 fields. best, Jon --_000_238089be2bf94c0298ba576eeb0cb647tcuedu_--