Since you're answering some questions, I have noticed a bug with times and dpm calculations, for example if you join a pickup as sub but only play 5 mins the logs still calculate dpm and heals per minute and all the time stats as if you played the whole 30 mins. Is that a bug in tftrue as well or is it a logs.tf bugs?

Ex: http://logs.tf/645676?highlight=76561197991320838

This is a bug on logs.tf.

zoob should answer this, but I think I read somewhere it will be fixed in the next big logs.tf update (see screenshots of it on page 4 of this thread).

Any ETA on the new profile pages or are they kaput?

I heard it will require an more powerful hosting solution and some sort of revenue stream needs to be identified.

Donation or subscription model maybe?

I would look into https://secure.dotabuff.com/plus as an example, they are kinda like the "logs.tf" of dota 2 and their monetization looks very effective.

At this point I think having ads would be understandable, then disable them for subscribers/donators.

buttnose was asking people if they were willing to pay £1 a month for a premium service on logs.tf. Don't know if there was follow up on that, and I have no idea why buttnose was doing it

Checking the viability of a donator mode I think, just as some research for zoob

What I would really like to see in a future update of logs.tf is 'Net Damage Output' which be total damage divided by damage taken.

Whilst most people automatically sort by damage given first, people often don't like at damage taken. For example if your pocket solider is putting out 390DPM but is taking 420DPM then his/her net damage output is -30DPM and therefore indicates that he/she is not helping contribute to his/her team as much as he/she could and soaking heals away from the rest of the team. I feel that net damage output is an important statistic to consider that most people don't necessarily look at. An individual like Gentleman Jon could have a field day with this.

People can quickly calculate this in their heads of course, but that also means the coding for this should be relatively simple and would encourage people taking this statistic into consideration, which in my opinion is quite important.

Disclaimer, i'm also well aware logs don't always accurately reflect in game performances.

TLDR: New statistic 'Net Damage Output' which could divide total damage by total damage taken (as well as per minute ) encouraging people looking at this IMO important statistic.

zoob, please look at some labels for damage dealt with specific unlocks if you can please.

It sometimes outputs strange stuff with loose cannon (damage labeled as grenade launcher and kills as lc), half zatoichi, shortstop, and some other unlocks i canr remember right now.

Thanks =)

Oh and also add "Ammo taken", because you know, if your Pocket took 80 ammo packs and did 6k damage, he's probably wasting rockets on nothing?

I mean, damage is damage... Ammo packs respawn anyway

This is a bug in TFTrue.

Kills while Ubercharged please.

oh? Okay, i didnt know.

Is it possible to implement a counter for Concherors used?

Good idea

Would be pretty cool and useful to have a search option similar to IMDB.com's collaborations and overlaps search. It lets you search for two actors/actresses/crew and gives you any movies/TV shows those people both worked on as the search results. So on logs.tf you would search for two player names or two steam ID's and the results would be any logs that both of those players played in.

I remember there being talks of that function being a reality when the site overhaul is done.

The 'map' field in the json api always seems to be empty even for logs that have the map shown on the site

I have two suggestions.

1) DPM should be based on the time a player spent playing and not the length of the entire match. Sometimes, I see, that if you join a game as a merc or a sub the time is listed as 0:00 even though you played for a few minutes. Though, other times you see the actual time played i.e. 4:51 of a 22 minute game. My suggestion is, in these instances, the DPM is worked out based on the 4:51 that you played and not the 22:00 of the entire game, which you were not there for. If you join a lot of games as a merc or subs in TF2Pickup and TF2Center, this would be great for you.

I understand this will only very rarely have any kind of meaning in official matches, but since probably less than 1% of logs uploaded to logs.tf are from officials, this would actually affect hell of a lot of logs.

2) Use ∞ instead of 0.0 for gamers who never died during the match, and sort them as the top (descending) gamer sorting by KD or KAD because they had a better KD or KAD than everyone else (unless others had not died either, and the sorting should reflect that.

Hey, you probably know about this already but logs.tf has been down for at least an hour due to a 500 HTTP error (Internal server error). Just thought I'd let you know in case you hadn't spotted it already. The fact that everyone is moaning about logs.tf being down just shows how much of an asset to the community it is. Keep up the good work! :)

http://i.imgur.com/g5CcSCn.png

????

Damn... That will be amazing timing for lan. Anything with more stats and graphs makes me feel warm and cozy

Fingers crossed for "real" dpm not counting overkill damage (looking at you, spies). And hopefully older logs can be viewed with whatever new features are being added!

I've always like that sizzling stats and logs differed in this way, and is actually one of the primary reasons I prefer logs. Just my opinion I guess

IMO, logs.tf's damage stats (except for spy, where the backstab would have killed them no matter what) are a more accurate representation of your quality of play, and sizzlingstats are a more accurate representation of how you played against a specific team alongside your team. If you go crazy hitting meatshots all over the place, but your opponents were already at low health, sizzlingstats would report that your damage was fairly low, as if you had been missing a lot of shots. It's not your fault that your opponent's health was low, and logs accurately reflect how much damage you *would have done* on shots that killed them.

I actually like logs better because of that

If a Scout has 10 health and a pipe him, I did 100 damage to him even if it only required 10 to kill him.

I do agree about the Spies backstab damage.

Also is there a possibility for stv integration like there is with sizzlingstats where you can go to the logs and have access to the stv if it was set up?

the ability to view both "raw damage" and "real damage" is possible and would make everyone happy :)

This is from a log i downloaded off of logs.tf

L 08/16/2015 - 16:37:45: "Nokk<25><[U:1:54758747]><Blue>" triggered "damage" against "goodnight sweet blingee :c<27><[U:1:67105874]><Red>" (damage "51") (realdamage "19") (weapon "scattergun")

interesting part is the last part

(damage "51") (realdamage "19")

Supplemental stats has implemented real damage for many months now. Any damage event that precedes a kill event also gives the real damage. Shouldn't be too hard to implement.

Any update?