Jump to content
Koumei & the Five Fates: Share Bug Reports and Feedback Here! ×
  • 1

A Web Based Dps Tool!


Question

Posted (edited)

The site is now updated for Damage 2.0!

 

The damage/level table has been replaced with the damage to specific enemy types (Ancient Disruptor, Grineer Napalm, Corpus Tech, and Corpus Moa), mouseovering the specific enemy type shows the vulnerabilities.

 

You can now also sort weapons based on damage to the enemy types (Ancient, Napalm, etc.).

 

Mouseovering the weapon stats table shows the original stats (for comparisons).

 

Not a big update feature-wise, but a ton of testing in-game and in the code. Next update will probably be about adding sentinel weapons, the one after that will likely be fixing page load times, and then perhaps adding more details in tooltips (any ideas are welcome).

 

Original post:

---

 

I've been using a spreadsheet to get a feel for how weapons perform, and how mods affect DPS. But it started becoming a bit of a pain to update/change mods around, and it felt kind of limited. So I tried to fix some stuff through scripting instead, using javascript and a regular web page (no downloading needed), and I thought you guys might get some use out of it too! Check it out (not sure if/how linking works on the forums):

 

http://dpsframe.com

 

Crash course in current features:

You select the pistol/rifle/etc. category you want at the top, to compare different types of weapons click "add favorite" and it'll get added on top of the other weapons (it remembers "favorites" through cookies, so it should get saved. As long as you stay on the same computer).

 

Increase/decrease aura/mod levels by using the + / - signs before the mod name. You can max or disable a mod completely by clicking the counter in the middle (for example the "10/10" for hornet strike on Acrid). A mod with a "-/5" as a counter is disabled, a mod with "0/5" is just the base card with no ranks added.

 

Mouse over the name of the mod to get a tooltip with what stats it increases.

 

The auras count up to 5, then start over from zero, so you can see how it stacks with more people. The number in parenthesis beside the counter (for example 5/5 (1) corrosive projection) represents how many players are need to get that high a percentage.

 

The DPS number to the right of the weapon name (for example "Acrid 6005") is the raw DPS value before armor is taken into account. The DPS with armor values is in the table just below and to the right of the name, it lists damage to Ancient Disruptors, Grineer Napalms, and Corpus Techs, levels 5 through 200. This damage does not account for hitting weak spots, but it does factor elemental weaknesses (fire/lightning/etc.) and armor ignore damage (armor piercing, serrated blade, poison, etc.).

 

You can sort the lists by clicking the names/arrows beside the category name, clicking name will sort by name, clicking it again will reverse the order. DPS means unmmitigated damage (doesn't factor armor), and APDPS means the damage after armor has been accounted for (using a lvl 200 Ancient Disruptor as base).

 

The "filter name" search box below favorites filters names, typing in "a" will show all weapon names that contains that letter (it's not case sensitive).

 

 

There's a ton of things left to work on:

There are no melee weapons currently (the swing mechanics seemed a bit complex).

 

The continuous (e.g flux rifle) and charge weapons (e.g Paris) just ignore weapon speed mods at the moment, which might not be accurate.

 

All stats have been entered by hand, so I can almost guarantee that there are typos or swapped stats.

 

I have checked the formulas and calculated some stats by hand, but there's still a few possible places where small errors might have snuck in.

 

Some special behaviors might not be modeled correctly, Acrid's DOT, Ogris explosion, Torid poison Cloud.

 

I haven't figured out a good way to represent the impact of recoil/spread. This might make some weapons seem very good on paper while in reality they might be horrible.

 

 

And some design limitations:

It's single target only.

 

It doesn't account for you being in a group. If you have a nova (2x damage from molecular prime), a Rhino (+50% damage shout), a Banshee with sonar (500% or more to sonar weak spot), and you're shooting a Moa in it's weak spot (fanny pack, 3x damage) with your synapse (electricity 2x damage) through your Volt shield (red crits), will do significantly more DPS than what's listed.

 

It's based on average DPS for emptying a clip and reloading. Burst DPS on some weapons will be higher, and this doesn't account for damage against shields. This misses things like that freezing damage might be better in some cases, or that you might attack a boss and then reload when he's invulnerable (e,g Vor, the burst DPS might be more important in some cases).

 

I tested in the latest versions of Chrome/Firefox/IE, I have no idea how good it'll work in anything other than those.

 

The javascript source is uglified, regular source code can be found on github:

 

https://github.com/Gogge/warframe-loadout

 

Some caveats on the code; It's almost completely uncommented and it's the very first iteration of a prototype.

Edited by Gogge

Recommended Posts

Posted

Don't forget you could fire such weapons using the scroll-wheel or a macro to maximise your dps.

scrollwheel :D genius :D never thought about that. :D

Posted

Just a heads up Pathogen Rounds on the website says it uses up 7 mod points but in the game it uses 11. By the way been loving this website, it has helped me out a lot

Posted (edited)

I have exams coming up this weekend (and a ton of other "this week only" limited betas that's soaking up my free time) so that's why updates are so slow.

 

As usual clear your browser cache if you don't see the new update.

 

12.2.1 Revision 6:

 

  • Updated burst weapon fire rate calculations (thanks Vanhline)

 

The delay between rounds in a burst was based on a 60 FPS test (0.0275 seconds), doing some more testing showed this to be affected by FPS. At 120 FPS the delay dropped to around 0.017-0.019 seconds, using 0.0185 seconds matches the calculations with in-game testing (see discussion in the posts below for details). I also removed the 0.00833 second delay between bursts, I'll probably refine the numbers further in future testing.

 

12.2.1 Revision 5:

 

  • Updated Castanas to have additional electric damage from the 100% electric proc (50% of base damage, not affected by elemental mods).
  • Updated Castanas to use Heated Charge (90% fire damage) instead of Ice Storm (40% cold damage and 40% magazine capacity).
  • Added Firestorm mod for Ogris/Penta/Ignis (doesn't affect DPS)
  • Fixed mastery rank for some weapons.

The 6 charges set at the same time limit on the Castanas makes a magazine size of 3 really awkward with multishot as you can proc 6 charges with only two uses. So you throw twice, then have to detonate, then throw once, and then detonate while reloading. With a magazine capacity of two you throw twice and detonate while reloading.

 

I can't see a really good way of modeling this behavior with the "maximum charges" weapons (like the Penta) without rewriting some things, so people will have to keep this in mind when looking at the DPS number if they choose to use increased magazine size and multishot mods at the same time. The default mods used now on the Castanas doesn't have this issue, so the DPS number is "fairly" representative of what you'll see in game (single target).

 

12.2.1 Revision 4:

 

  • Moved Grinlok from the sniper category to the rifle category (thanks Telogor and MeduSalem)

 

12.2.1 Revision 3:

 

  • Fixed Castanas projectile count typo.

 

12.2.1 Revision 2:

 

  • Fixed Rifle Amp aura typo for void DPS.

 

Updated for 12.2.1!

 

 

  • Added the Grinlok (based on Codex stats)
  • Added the Castanas (based on Codex stats)
  • Updated burst weapon handling (thanks GottFaust)
  • Changed the shotgun proc chance to reflect status mechanic changes (cosmetic, status procs are still not implemented)
  • Updated Phage stats to match codex (1.0 fire rate, thanks Andoryuu and others)
  • Updated Pathogen Rounds (pistol toxin mod) with proper base mod cost (thanks Kirito_Okazaki)
  • Changed weapon initialization to reduce initial load times slightly

 

 

The burst weapon changes reflect that each bullet fired has an individual delay (like 0.06 seconds for the Hind), which helps low fire rate weapons but slightly hurts high fire rate weapons. As an example the base Hind has a fire rate of 5.0 (5 shots per second, or 0.2 seconds between each shot), and a burst delay of 0.0875 seconds (0.06 for weapon and a 0.0275 default delay). So it takes 0.6375 seconds to fire five shots (5 * 0.0875, and 0.2 for the fire rate delay between bursts), if you just fired five shots regularly without burst it'd take a whole second (meaning burst is more DPS).

 

If we add Speed Trigger and Shred we increase the fire rate to 9.5, and it takes ~0.54276 seconds to fire five shots (5 * 0.0875 + ~0.105). If we just used auto fire it'd only take ~0.52632 seconds (5 * ~0.105). So in this second case burst fire is less effective than regular auto fire would have been. In the calculations I also add a second delay between each burst (after the 5 shots in the case of the Hind) of 0.00833 seconds based on in-game footage, but this might just be because the Hind is semi-automatic and the rapid fire macro might be FPS limited (I recorded at 120 FPS, which is.. 0.00833.., amazing coincidence).

 

I added a "Fire Rate (burst)" stat to burst weapons that shows this effect (fire rate mods are still a DPS increase, just slightly less effective at higher fire rates).

Edited by Gogge
Posted

Err, Grinlok and Castanas aren't in the listing, but thank you for updating!!

We missed you :3

 

Sometimes the browser caches the javascript (calculations/weapons) and shows the updated html (menus/etc.), making it seem like it's the updated version but missing some things (like weapons). Clearing the browser cache makes it download the new javascript and makes the new weapons appear (usually the browser cache expires eventually and you get the new version anyway).

 

I'll see if I can put the version number in the javascript instead and it should be more noticeable if this happens.

Posted (edited)

But it's not a Sniper rifle as it draws from normal rifle ammo, so I guess it's missplaced?

 

Yeah, having only 6 rounds in the magazine I had assumed (incorrectly) that it was a sniper rifle when I added it. I've moved it to the rifle category, thanks.

Edited by Gogge
Posted

Yeah, having only 6 rounds in the magazine I had assumed (incorrectly) that it was a sniper rifle when I added it. I've moved it to the rifle category, thanks.

 

It probably should be sniper, because having 540 ammo on the Grinlok is just basically a Sniper with infinite ammo, as it inflicts as much damage as most of the snipers do. It also has the highest damage per hit of any rifle available currently. But that's another mark of uncorrectly balanced weapons in DE's history. :P

 

Thanks for updating it!

Posted (edited)

There's no firestorm mod for Penta available.

 

I usually don't include the mods that doesn't affect single target DPS, unless people request them. I've added the Firestorm mod to the Penta/Ogris/Ignis mod lists.

Edited by Gogge
Posted

im questioning the information regarding the burst fire mode delays. with my burston prime i do notice that slight delay when i press the button manually but that is caused by the rate i spamming the fire button is not matching up with the fire rate delay of the weapon and is whats causing the delay. when i use a fire macro that clicks ever millisecond and means fire is being clicked THE INSTANT the previous burst ends and removes user delay i am able to hit the 16 rounds per second without a issue, not the 10.98 that the change suggests. 

 

it seems to me that the delay being added is user generated and not in fact a limitation on the rifle.

Posted (edited)

im questioning the information regarding the burst fire mode delays. with my burston prime i do notice that slight delay when i press the button manually but that is caused by the rate i spamming the fire button is not matching up with the fire rate delay of the weapon and is whats causing the delay. when i use a fire macro that clicks ever millisecond and means fire is being clicked THE INSTANT the previous burst ends and removes user delay i am able to hit the 16 rounds per second without a issue, not the 10.98 that the change suggests. 

 

it seems to me that the delay being added is user generated and not in fact a limitation on the rifle.

 

The main limiting factor is the 0.0275+0.04 seconds delay between each bullet fired in the three round bursts which isn't affected by user interaction. The user-affected 8.33 ms delay at the end of the burst only drops the fire rate by 0.3 (10.98 vs. 11.32 with Speed Trigger and theoretical 16 fire rate).

 

Going back and re-testing it and checking frames between each shot fired in the burst (on a Hind) I get 6->5->5->5 frames (0.1->0.0833.. seconds) at 60 FPS, and 5->5->4->5 or 5->5->4->4 with 120 FPS, so it might be yet another mechanic related to frame rate. As I haven't normalized the Stug/Akbolto to 60 FPS it might be appropriate to drop the 0.0275 default delay between shots for the burst weapons to 0.0185 (what I get at 120 FPS, not sure how low the delay gets when you get the frame rate even higher).

 

This gives the Hind with Speed Trigger and Shred a fire rate of 10.04 (recording at 60 FPS and checking frames in-game at 120 FPS it's 10.02), and the Burston 11.96 (in-game 11.95). I'll have to grind the Burston Prime to test it, not sure when I'll get the time.

Edited by Gogge
Posted

The main limiting factor is the 0.0275+0.04 seconds delay between each bullet fired in the three round bursts which isn't affected by user interaction. The user-affected 8.33 ms delay at the end of the burst only drops the fire rate by 0.3 (10.98 vs. 11.32 with Speed Trigger and theoretical 16 fire rate).

 

Going back and re-testing it and checking frames between each shot fired in the burst (on a Hind) I get 6->5->5->5 frames (0.1->0.0833.. seconds) at 60 FPS, and 5->5->4->5 or 5->5->4->4 with 120 FPS, so it might be yet another mechanic related to frame rate. As I haven't normalized the Stug/Akbolto to 60 FPS it might be appropriate to drop the 0.0275 default delay between shots for the burst weapons to 0.0185 (what I get at 120 FPS, not sure how low the delay gets when you get the frame rate even higher).

 

This gives the Hind with Speed Trigger and Shred a fire rate of 10.04 (recording at 60 FPS and checking frames in-game at 120 FPS it's 10.02), and the Burston 11.96 (in-game 11.95). I'll have to grind the Burston Prime to test it, not sure when I'll get the time.

 

 

I'd be more than happy to give you Burston Prime parts if you want, just pm me.

Posted

As usual clear your browser cache if you don't see the new update.

 

Updated to 12.3.0:

  • Added Marelok.

 

Vorigan kindly donated the Burston Prime parts so I'll do some testing on the Burston this weekend hopefully.

Posted (edited)

Is there a way to change the default mods?

 

Or atleast add another default mod list for a lvl 30 base weapon with 30 mod points that would be a common setup for people.

Edited by Vindicit
Posted (edited)

Is there a way to change the default mods?

 

Or atleast add another default mod list for a lvl 30 base weapon with 30 mod points that would be a common setup for people.

 

There's no way to change the default mod setup at the moment, adding a way for people to set it manually for all weapons would require a bit of work (I could perhaps add something like a "copy mods to other weapons"-button, I'll have to think about it).

 

One of the problems with adding a 30-point setup is that there's no default setup that works for all weapons, and making individual 30-point setups for the weapons would be a bit of work.

Edited by Gogge
Posted

I really, really gotta ask you man.

Have you considered doing melee yet?

It would be amazing if you would, I know its harder to tell dps than with weapons, but anything would be hugely appreciated.

Posted

I really, really gotta ask you man.

Have you considered doing melee yet?

It would be amazing if you would, I know its harder to tell dps than with weapons, but anything would be hugely appreciated.

 

I've been thinking about it, but I haven't looked at how melee works (it'll take time to implement), and from comments by others it seems like melee normal/charge attacks have different stats so there's effectively double the weapons to add (it'll take time to add). We also have melee 2.0 coming so that might change things a bit, implementing the system before that might just mean I'd have to re-write things for 2.0 (which would take more time).

 

It's on the todo list, but I have no guess as to if/when it might be added.

Posted (edited)

As usual clear your browser cache if you don't see the new update.

 

12.4.0 revision 2:

  • Updated Accelerated Blast to only affect Puncture damage
  • Updated the default Shotgun mods (due to the Accelerated Blast fix)

The Accelerated Blast change makes it behave like other IPS mods (in game tests show the same as codex numbers), it only increases the puncture based portion of the base damage (not sure if it still adds puncture to Phage in game).

 

Updated for 12.4.0!

  • Added the Boltor Prime (based on codex)
  • Added the Akbronco Prime (based on codex)

Based on the codex numbers the Boltor Prime seems a bit overpowered, but DE might be factoring that it's using projectiles and that it has a fairly low status chance (still higher than the Soma tho). The Akbronco Prime seems nice, the status procs will bring up the DPS a bit in game compared to the listed DPS (I should really fix so the page calculates status proc damage).

Edited by Gogge
Posted

dpsframe hasn't been playing so well with my Chrome. Any reasons why it is so laggy compared to IE? :(

 

I almost exclusively use Chrome (both for developing and regular use), so I have no idea (addons?). I get no noticeable difference between the two when testing.

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
×
×
  • Create New...