Go Back   Star Trek Online > Feedback > Builds, Powers, and Game Mechanics
Login

Reply
 
Thread Tools Display Modes
Lt. Commander
Join Date: Dec 2007
Posts: 120
# 91
04-14-2012, 02:45 PM
Can someone post a link to the download? I can't seem to find the DPS meter thing, and it looks like all of the websites that had it took it down...

Thanks,
Quiiliitiila
Lt. Commander
Join Date: Dec 2007
Posts: 120
# 92
04-14-2012, 03:54 PM
Quote:
Originally Posted by Quiiliitiila
Can someone post a link to the download? I can't seem to find the DPS meter thing, and it looks like all of the websites that had it took it down...

Thanks,
Quiiliitiila
http://forums.startrekonline.com/sho...5&postcount=67
Search is your friend

Quote:
Originally Posted by kostamojen View Post
You aren't accounting for the power sucking abilities of well built Sci/Carrier builds.

My fastest Infected Elite run ever had 1 carrier and 4 escorts. Had both transformers dead with more than 10 minutes left on the clock, and the tac cube lasted less than 30 seconds.

They key was the Carriers ability to suck ALL the power from a target, so NPC's like the normal cubes and tac cube (even the Gate) simply lost all their shields and just sat there unable to fire, allowing all the Escorts to hit the targets with their full DPS. The loss of the 5th Escort didn't matter, because even with 5 escorts you still won't be able to burn through the larger targets that quickly, and I've had dozens of all-Escort runs to back that up.

Now I have no idea how to perfect those power-drain builds, but every time I run a STF with someone using one of those setups its apparent how effective they are in improving the speed at which an STF can be completed.
First of all, posting times here isn't really relevant, I've killed the tac cube in 20 seconds and finished an optional with almost 12 minute on the clock. What does that mean? It means nothing, it means I had a bunch of tryhards go in with the specific goal to make their e-peen bigger.

2nd, I actually did address power siphoning and cc.
To summarize again, cc makes everyone's life easier, but it does not increase speed because 5 Escorts do not require cc to kill a tactical cube.
There is no loss to DPS because no one dies or is required to do anything but shoot and in emergencies, throw a heal on whoever has aggro, although personally I have not had that happen yet. 1 Escort can tank a tactical cube for at least 30 seconds at which point it's long since dead.

It is completely safe to just have 5 escorts sit there and spam their buttons at a tactical cube, it will mathematically go down faster barring a heavy torpedo hitting someone. Power siphon drones don't save you from that.

I'm not saying that having a science ship isn't useful or that it doesn't have it's place. Indeed if you read the entire post you'll note that I fully support science vessels.

I'm saying that mathematically it is faster to have 5 escorts and saying that having a science vessel increases your speed is incorrect.
cc science vessels decrease difficulty. I would much rather have 1 cc science vessel and 4 escorts because even though I may optimally shave off a few minutes of time I know I wont have to worry about pressing my defensive buttons.
Lt. Commander
Join Date: Dec 2007
Posts: 120
# 93
04-14-2012, 04:04 PM
Quote:
Originally Posted by ekental View Post
Ah, thanks. I didn't think to search the forums, only the web.

-Quiiliitiila
Lt. Commander
Join Date: Dec 2007
Posts: 120
# 94
04-15-2012, 03:30 AM
Quote:
Originally Posted by ekental View Post
1 Escort can tank a tactical cube for at least 30 seconds at which point it's long since dead.
If you have a video of yourself tanking the final Tac Cube in ISE in your escort for 30+ secs, I'd love to see it to get an idea of the CDs, as I usually get obliterated quite fast (Tac/Fleet Escort).
Lt. Commander
Join Date: Dec 2007
Posts: 120
# 95
04-15-2012, 04:49 AM
BTW I managed to run the combat parser (version 0.1, version 0.2 just gives me an empty report window).

You should be very careful with the DPS values:

- it's missing stuff, the totals I get are lower than the ones I get from STOICs and from my own log parser (which closely matches STOICs. BTW). I don't think it's a problem with the source/owner thing (which should be fixed by setting the right option), it could be caused by the malformed lines which miss the attack type (but have the right internal type). Before you ask: the combat log provided by STO is barely decent, as it's missing a ton of important information. It's not a surprise for a combat logger which is some years old to be wrong.

- it does not give you the DPS value: the usual name for that value is DPS(e) which means "dps over the duration of the encounter", which is calculated as total damage / total encounter duration. Since the combat log is range-limited, this will overvalue your DPS(e) compared to that of the other players. A somewhat better approach is to use the real DPS, which is damage done / time spent in combat, that is less affected by the range problems (ally out of range => you see no damage, but you also don't see him in combat. The final result will still be wrong, but less than with DPS(e), if you really want reliable results, everyone should record a log and you should use each player's log to determine their DPS).
Note also that using DPS(e) instead of DPS makes it impossible to compare players unless the total duration of the instance was the same: for the same player (and damage output) a fast run with an organized group will always provide a higher DPS(e) than a pug group hitting the optional objective with 5 seconds left.
Lt. Commander
Join Date: Dec 2007
Posts: 120
# 96
04-15-2012, 05:18 PM
Quote:
Originally Posted by pezacoatl
BTW I managed to run the combat parser (version 0.1, version 0.2 just gives me an empty report window).

You should be very careful with the DPS values:

- it's missing stuff, the totals I get are lower than the ones I get from STOICs and from my own log parser (which closely matches STOICs. BTW). I don't think it's a problem with the source/owner thing (which should be fixed by setting the right option), it could be caused by the malformed lines which miss the attack type (but have the right internal type). Before you ask: the combat log provided by STO is barely decent, as it's missing a ton of important information. It's not a surprise for a combat logger which is some years old to be wrong.

- it does not give you the DPS value: the usual name for that value is DPS(e) which means "dps over the duration of the encounter", which is calculated as total damage / total encounter duration. Since the combat log is range-limited, this will overvalue your DPS(e) compared to that of the other players. A somewhat better approach is to use the real DPS, which is damage done / time spent in combat, that is less affected by the range problems (ally out of range => you see no damage, but you also don't see him in combat. The final result will still be wrong, but less than with DPS(e), if you really want reliable results, everyone should record a log and you should use each player's log to determine their DPS).
Note also that using DPS(e) instead of DPS makes it impossible to compare players unless the total duration of the instance was the same: for the same player (and damage output) a fast run with an organized group will always provide a higher DPS(e) than a pug group hitting the optional objective with 5 seconds left.
If you have a better parser, you should consider sharing it so everyone can benefit from the increased accuracy.

I am aware that sometimes there are inconsistencies. That being said, comparing the Internal Log in a spreadsheet to the CombatLogParser, I've found the parser to be within a 5% accuracy range. As for damage sources, CombatLogParser 0.1 by default sums all source damage together and credits it to the owner.
i.e. If a science captain hits Photonic Fleet, the sum damage of that fleet will be added to his damage. Clicking on the outgoing damage child of the captain would show damage sources that they don't even have

If you right click on the encounter and click "Credit Dmg by Source" you can get an individualized breakdown. I do however, think that summing that damage up is the correct method.
Additionally, you'll note that the CombatLogParser by deault doesn't report shield damage, which is where major discrepancies occur and could also be where your discrepancies originate from.
Part of this because shields don't seem to count for aggro, and another part is because counting shield damage may possibly skew your damage numbers as shields have different resists than hull in PvP. (PvE nothing has resists from only Borg STF tests)

If you refer to the original post made on page 7, I have a specific paragraph talking about the problems with logging in general, DPS, time, etc. You'll notice that I do not advocate simply /CombatLog 1 and letting it run for the entire STF.

When I use /CombatLog, I try to log long fights, where buffs designed to spike dmg for a short time are minimzed so that I can get a good feel for what people's sustained DPS is. That usually means a boss fight, or even a generator if I just want to see how high the numbers can get.
This gives me a potential DPS, and then the rest of the equation is just trying to get as close to that value as possible.
Internal Combat Logging is actually really good about stopping the log as soon as combat is over. That being said I also manually /CombatLog 0 at the end of a fight in case residual dots are draining someone's DPS by extending time. Even then, I assume a 10% DPS variation in everyone's numbers due to mistimed rotations etc.

Finally, I haven't extensively tested range issues, but the only time it may have made a difference in STFs is the defenders in KASE and CSE. I have yet to see a major discrepancy caused by someone going out of range, and it has been a major worry for me since I started using /CombatLog. It may be an issue with settings, I haven't looked into it. Still, as previously noted, I don't often log entire STFs because the differing roles and strategies usually result in a lack of useful information.

Side Note: Will see about a video, I need to update encoding tools first T_T.
Lt. Commander
Join Date: Dec 2007
Posts: 120
# 97
04-16-2012, 09:10 AM
Quote:
Originally Posted by ekental View Post
If you have a better parser, you should consider sharing it so everyone can benefit from the increased accuracy.
That was the idea from the start. I have a friend testing the latest version, which can actually be controlled by buttons instead of vodoo magic and the command line. I'll then clean the source and ready it for distribution (it's done using Qt4/free, which means GPL, which means you get the source code).

Quote:
I am aware that sometimes there are inconsistencies. That being said, comparing the Internal Log in a spreadsheet to the CombatLogParser, I've found the parser to be within a 5% accuracy range. As for damage sources, CombatLogParser 0.1 by default sums all source damage together and credits it to the owner.
Since the total damage is just a sum of fields in the combat log, the match should be exact. At least if the totals stick to the only reliable data in the log, which means:
- damage fired (i.e. what your ship spits out)
- hull damage done (i.e. how many hull HP the target lost)

All the rest is lost in the misty haze of missing data..... (example in case: shield drains). Shield damage can be semi-guessed, any serious attempt to account for it would be a lot of work (and probably not 100% reliable).

Quote:
Additionally, you'll note that the CombatLogParser by deault doesn't report shield damage, which is where major discrepancies occur and could also be where your discrepancies originate from.
Part of this because shields don't seem to count for aggro, and another part is because counting shield damage may possibly skew your damage numbers as shields have different resists than hull in PvP. (PvE nothing has resists from only Borg STF tests)
Yes, this could be the explanation. For my reports I provide both damage fired and damage done, with anything not going to hull damage being considered shield damage. Even if the values are approximate, I still find shield damage to be relevant.

Quote:
When I use /CombatLog, I try to log long fights, where buffs designed to spike dmg for a short time are minimzed so that I can get a good feel for what people's sustained DPS is. That usually means a boss fight, or even a generator if I just want to see how high the numbers can get.
This basically means "never".... What we would need is a target dummy, i.e. a cube or something with infinite hull (or shields) that a group can fire on for some time to compute the DPS. I remember reading on the wiki that there's a passive cube in some of the episode missions, but I'm not there yet.

BTW for me the main aim was to have my damage spliit by ability, to see which weapons and skills are doing the most damage. It's only when I started to find that I was very (too) often top DPS/damage that I started to question the validity of the data. I may be good, but I'm definitely not THAT good. Running together with a friend and comparing the logs gives you an idea of how much things can go wrong. I've seen a deviation of 30%, but last time we stuck together and the difference was a lot smaller.
My program only cares about damage done/dps, and full statistics are given at the end, covering the entire log, and provides tools to manage your logs (backup, save statistics, etc.), the idea is that you run it for the duration of the combat/instance you care about and then move the log out of the way to avoid ending up with a file of 1Gb.... this means no splitting in separate combats, the totals are always for the full log.

Quote:
Side Note: Will see about a video, I need to update encoding tools first T_T.
Hehe, same here
Lt. Commander
Join Date: Dec 2007
Posts: 120
# 98
04-16-2012, 04:05 PM
Quote:
Originally Posted by drasketo
new and worse death penalties coming in a few days.
What?!
/10char
Lt. Commander
Join Date: Dec 2007
Posts: 120
# 99
04-16-2012, 08:54 PM
Quote:
Originally Posted by pezacoatl
If you have a video of yourself tanking the final Tac Cube in ISE in your escort for 30+ secs, I'd love to see it to get an idea of the CDs, as I usually get obliterated quite fast (Tac/Fleet Escort).
I do it myself as well regularly. The important parts are as follows:
2x Uni resist consoles, 1x kinentic
Maco shield/deflector (i use mk 10)
borg engines/console/shield cap thingie
subspace device that boost resists

Boff abilities
2x E power to shields
1x Aux 2 Struc *this is an under-rated power IMHO*
1x Hazard Emitters
1x TSS
2x Tac Team

Skills: Max all that apply to hull/shield capacity & healing, and power insulators, at least 6 in shield power, warp efficiency/capacity, and at least 3 in aux power. Only need 3 in armor skills and those aren't really required.

Fly above cube and down on top of it point blank throttle at 1/2 or 1/4. Rotate boff abilities as common sense dictates (IE don't blow HE or TSS until you can gain mostly the full heal benefits, use others the second they are up). This can tank it for far longer than 30 seconds as well if needed. If you need to do a 'super tank' mode simply switch to nearly full aux power rest in shields go full throttle and circle strafe it, with your defense high you can tank it pretty much indefinitely barring a nasty torp hit then switch back to offensive while it pops a cruiser.
Lt. Commander
Join Date: Dec 2007
Posts: 120
# 100
04-17-2012, 01:12 PM
Here is a link to the old STOIC parser
http://malicia.straygenius.com/wiki/...ombat_Software

Not sure how well it works now as the coder for it hasn't been seen in a while in STO.
So it may well be out of date.
Reply

Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is Off
HTML code is Off


All times are GMT -7. The time now is 11:58 AM.