Doom 2016 movement speed is 230 Units Per Second

[MuscularMelvin]

People have been arguing over whether the new Doom is faster or slower than the classic ID shooters. Purists deride it as console trash, whereas naive millenials praise it as being even faster. I decided to log just how fast the player moves in Doom 2016 using MAP01's hallway as the benchmark. As it turns out...

Doom 2016 movement speed is 230 units per second.

• Doom II - 1.5 seconds - 583.33ups
  
• Q3 Haste - 2 seconds - 400ups
  
• Quake 1/3 - 2.5 seconds - 320ups
  
• Quake 2 - 2.8 seconds - 300ups
  
• Doom 3 MP - 3.1 seconds - 270ups
  
• Doom 2016 - 3.6 seconds - 230ups
  
• Doom 3 SP - 3.7 seconds - 220ups

I calculated this using Doom II's MAP01, Matthias Worch's MAP01 remake[www.quaddicted.com] for Quake 1, and Doom 2016. I used a stopwatch to time how fast I could get from one end of the hallway to the other (with the RL secret closed) and logged the results. In addition, Doom 2016's strafing speed is also 230ups and backwards speed is 205ups. Walking/crouching is 88ups. Also, Haste running speed seems to be around 300ups because in the test I conducted, the Haste lap time was 0.8 seconds faster than the default speed.

As the results clearly show, Doom 2016 is not the super-fast game that fans had originally thought it was. For all the complaints levelled against Doom 3's multiplayer, it appears that even that game has faster movement than Doom 2016 does.

I know this will be lost on deaf ears on the Steam forum. I'll have to make a Doomworld account and repost this there, sometime.

Technical explanation:the multiplayer beta had a value known as pm_runspeed that showed that the Doom Slayer ran at 400UPS. However, it turns out that the player size in the IDTech 6 Engine is much larger than in any previous IDTech engine game. The MAP01 hallway is 864 units long in Doom II, but the Doom 2016 remake's version is 1695.5 units long. Both versions of the map appear identical when both games are set to the same FOV (e.g. if Doom II = 100FOV, then Doom 2016 = 115FOV).

[MuscularMelvin]

I might also mention that viewheight is separate from player heights. Doom 2, for example, places the camera at chest level.

[budozero]

My point is that the factual figures mean nothing if perception is not taken into account. This is the basis of what you started your research based on, what people "felt", is it not? I dont think the factural figures clarify it.

引用自 volcanic_lightning：

I might also mention that viewheight is separate from player heights. Doom 2, for example, places the camera at chest level.

Most games have a different view hight compared with player height. We can thank this for head glitching. I recall the UT3 beta had this to an extreame where you could see over objects that completely obscured you from view.

[Comp_Lex]

引用自 volcanic_lightning：

No, I'm not missing the point. The point of this topic was to calculate the movement speed in Doom 2016 and compare it to the values in classic ID games. It was not to analyze the speed of the gameplay itself. You are the one who decided to come into this purely objective topic about hardcoded movement speed values in retro FPS games and preach about your own subjective ideas on how Doom 2016 feels and plays.

The classic maps are the same size as in the original games; they're just scaled up for the player's size in Doom 2016. The FOV values are the same as well; some games convert 4:3 HFOV values and some, like Doom 2016, do not.

And @budozero, IDTech engines up until IDTech 5 use the exact same map units. The Source engine was built upon GoldSRC, which itself was forked from IDTech 1.

No, no, no. That's what you did ===> "As the results clearly show, Doom 2016 is not the super-fast game that fans had originally thought it was."
It actually is a super-fast game. You cannot base such a conclusion on a single metric. You have to take into account the entire game (enemies, level design, weapons, etc.)

[MuscularMelvin]

Actually, yes. Yes I can. The Doom 2016 player character is objectively slower than the player characters in classic ID games. I proved it by timing how fast Doom Slayer could run down the MAP01 hallway, then I calculated that speed by messing with Quake 1's cl_forwardspeed and cl_maxvelocity parameters in a 99% accurate recreation of MAP01.

What you feel is completely irrelevant. By that logic, I can say Doom 2016 is a slow game because I feel that the classic maps play really slow and clunky compared to how they play in the original games. But that is not true; the enemy balance is incompatible, so the comparison is moot. What isn't moot however is the hardcoded base running speed between games. In that particular sense, Doom 2016 is slower than any of the previous ID games.

Since you're so insistent on tauting your opinions on Doom 2016 as fact in a topic that's about the objective coding behind these games, I'll throw you a curveball: Haste. I decided to go back and calculate Haste movement speed. The result was that Haste was faster than default running speed by 0.8 seconds. Going by the chart in the OP, that means Haste movement is around 300ups. I feel that Doom 2016 would've played much better if Haste movement was the default running speed, but I'm sure you're already preparing a long-winded essay on how that would ruin the balance of the game or whatever.

One last thing: I will admit I was wrong about the TF2 Heavy. Unlike GoldSrc, it seems Source engine movement speeds aren't the same as in IDTech. The E1M1 recreation I tested was quite inaccurate, but if I were to take a guess, I would say the Doom Slayer is a little faster than the Spy, but still much slower than the Scout.

[Ransom]

just a question: how big is 1 unit?

[I'll swallow your soul!]

引用自 Comp_Lex：

引用自 volcanic_lightning：

No, I'm not missing the point. The point of this topic was to calculate the movement speed in Doom 2016 and compare it to the values in classic ID games. It was not to analyze the speed of the gameplay itself. You are the one who decided to come into this purely objective topic about hardcoded movement speed values in retro FPS games and preach about your own subjective ideas on how Doom 2016 feels and plays.

The classic maps are the same size as in the original games; they're just scaled up for the player's size in Doom 2016. The FOV values are the same as well; some games convert 4:3 HFOV values and some, like Doom 2016, do not.

And @budozero, IDTech engines up until IDTech 5 use the exact same map units. The Source engine was built upon GoldSRC, which itself was forked from IDTech 1.

No, no, no. That's what you did ===> "As the results clearly show, Doom 2016 is not the super-fast game that fans had originally thought it was."
It actually is a super-fast game. You cannot base such a conclusion on a single metric. You have to take into account the entire game (enemies, level design, weapons, etc.)

Slightly faster than your average game of today on the market doesn't mean "super-fast game".

[Salamand3r-]

引用自 volcanic_lightning：

Actually, yes. Yes I can. The Doom 2016 player character is objectively slower than the player characters in classic ID games. I proved it by timing how fast Doom Slayer could run down the MAP01 hallway, then I calculated that speed by messing with Quake 1's cl_forwardspeed and cl_maxvelocity parameters in a 99% accurate recreation of MAP01.

What you feel is completely irrelevant. By that logic, I can say Doom 2016 is a slow game because I feel that the classic maps play really slow and clunky compared to how they play in the original games. But that is not true; the enemy balance is incompatible, so the comparison is moot. What isn't moot however is the hardcoded base running speed between games. In that particular sense, Doom 2016 is slower than any of the previous ID games.

Since you're so insistent on tauting your opinions on Doom 2016 as fact in a topic that's about the objective coding behind these games, I'll throw you a curveball: Haste. I decided to go back and calculate Haste movement speed. The result was that Haste was faster than default running speed by 0.8 seconds. Going by the chart in the OP, that means Haste movement is around 300ups. I feel that Doom 2016 would've played much better if Haste movement was the default running speed, but I'm sure you're already preparing a long-winded essay on how that would ruin the balance of the game or whatever.

One last thing: I will admit I was wrong about the TF2 Heavy. Unlike GoldSrc, it seems Source engine movement speeds aren't the same as in IDTech. The E1M1 recreation I tested was quite inaccurate, but if I were to take a guess, I would say the Doom Slayer is a little faster than the Spy, but still much slower than the Scout.

Since there is no universal metric, the only valid comparison is the subjective "feel" of the game.

Something may move 1 unit/s if those units are analogous to feet, 12 units/s of the units are analogous to inches, 30 units/s if they are analogous to centimeters.

There is no coherence between classic Doom, Quake 1/2, and modern idTech games. You keep trying to imply that all idTech games are the same in the definition of units, but a small amount of reading on mod forums indicate that they are not.

You are trying to reconcile incompatible systems. It's like fitting Star Trek canon into the Star Wars universe. Most of one doesn't translate to the other, and seldom makes sense when it does.

You did some interesting work, to be sure, but since the entire premise is false - that idTech games all define units identically - it is meaningless. Someone brough up a Halo speed comparison already. That is no more or less valid than your comparison, because units is an arbitrary measurement.

[Ransom]

引用自 Salamand3r：

引用自 volcanic_lightning：

Actually, yes. Yes I can. The Doom 2016 player character is objectively slower than the player characters in classic ID games. I proved it by timing how fast Doom Slayer could run down the MAP01 hallway, then I calculated that speed by messing with Quake 1's cl_forwardspeed and cl_maxvelocity parameters in a 99% accurate recreation of MAP01.

What you feel is completely irrelevant. By that logic, I can say Doom 2016 is a slow game because I feel that the classic maps play really slow and clunky compared to how they play in the original games. But that is not true; the enemy balance is incompatible, so the comparison is moot. What isn't moot however is the hardcoded base running speed between games. In that particular sense, Doom 2016 is slower than any of the previous ID games.

Since you're so insistent on tauting your opinions on Doom 2016 as fact in a topic that's about the objective coding behind these games, I'll throw you a curveball: Haste. I decided to go back and calculate Haste movement speed. The result was that Haste was faster than default running speed by 0.8 seconds. Going by the chart in the OP, that means Haste movement is around 300ups. I feel that Doom 2016 would've played much better if Haste movement was the default running speed, but I'm sure you're already preparing a long-winded essay on how that would ruin the balance of the game or whatever.

One last thing: I will admit I was wrong about the TF2 Heavy. Unlike GoldSrc, it seems Source engine movement speeds aren't the same as in IDTech. The E1M1 recreation I tested was quite inaccurate, but if I were to take a guess, I would say the Doom Slayer is a little faster than the Spy, but still much slower than the Scout.

Since there is no universal metric, the only valid comparison is the subjective "feel" of the game.

Something may move 1 unit/s if those units are analogous to feet, 12 units/s of the units are analogous to inches, 30 units/s if they are analogous to centimeters.

There is no coherence between classic Doom, Quake 1/2, and modern idTech games. You keep trying to imply that all idTech games are the same in the definition of units, but a small amount of reading on mod forums indicate that they are not.

You are trying to reconcile incompatible systems. It's like fitting Star Trek canon into the Star Wars universe. Most of one doesn't translate to the other, and seldom makes sense when it does.

You did some interesting work, to be sure, but since the entire premise is false - that idTech games all define units identically - it is meaningless. Someone brough up a Halo speed comparison already. That is no more or less valid than your comparison, because units is an arbitrary measurement.

100% correct :)

[MuscularMelvin]

You did some interesting work, to be sure, but since the entire premise is false - that idTech games all define units identically - it is meaningless.

You fail to understand that 230ups is not the value for Doom 2016--it is in fact the value for Quake 1, because that's the speed value I actually had to reduce it to in order to reach the end of the MAP01 hallway in 3.6 seconds. The scale of the maps to the player size is nearly identical between all 3 maps. The only discrepancy is that Q1's recreation was several units larger than Doom 2, but that number is too small to affect the lap times in any way.
Doom II at 100 FOV.[cdn.discordapp.com]
Doom 2016 at 115 FOV.[cdn.discordapp.com]
The hardcoded value for Doom 2016 is actually 400 if the multiplayer beta value is still used in the final.

[Salamand3r-]

引用自 volcanic_lightning：

You did some interesting work, to be sure, but since the entire premise is false - that idTech games all define units identically - it is meaningless.

You fail to understand that 230ups is not the value for Doom 2016--it is in fact the value for Quake 1, because that's the speed value I actually had to reduce it to in order to reach the end of the MAP01 hallway in 3.6 seconds. The scale of the maps to the player size is nearly identical between all 3 maps. The only discrepancy is that Q1's recreation was several units larger than Doom 2, but that number is too small to affect the lap times in any way.
Doom II at 100 FOV.[cdn.discordapp.com]
Doom 2016 at 115 FOV.[cdn.discordapp.com]
The hardcoded value for Doom 2016 is actually 400 if the multiplayer beta value is still used in the final.

Here's the thing - those maps were created to "feel" right in each game. They aren't ports or conversions, they are wholly different maps, each built from the ground up to be playable in the engine they were designed for.

There is still no valid comparison, since you are comparing arbitrary metrics mixed with artistic license.

You literally cannot make the comparison objectively.

[∙]

Any info on model height and viewheight in Doom2016? I know in Quake3 those are: player height=56units, 1unit=12 real life ft and if not mistaken viewheight=32units

[MuscularMelvin]

The map is the same size relative to the player in each game. That's why I showed the screenshots. How is that not a valid comparison?

Doom 2016 movement speed, when converted to the units used in Quake 1, would only be 230ups. Doom II movement speed, when converted to the units used in Quake 1, is around 583ups; the same value people on the Doom Wiki were able to calculate.[doomwiki.org] If the maps were created to "feel right in each game", why would the results be almost identical in every test I conducted between the 3 games?

[LemonRaptor]

i could easily criticise you're "method" of coming to that theory... but instead i will humor you. even if the movement is slower...who cares how fast it is ? it matters how fast it feels.

something that feels fast but is not fast is functionally the same as something that is fast, for the person playing.

same as heights between games, one height in one game may be far higher, whilst the other is lower, functionally the higher one could easily be the same as the lower one because it gives the same feeling to the player, it matters who designs it better to feel high, not how high it is.

[MuscularMelvin]

I'm not saying Doom 2016 is not a fast-paced game.

I'm just saying that the base movement speed in Doom 2016 is far slower than in previous ID FPS titles.

[Ransom]

引用自 volcanic_lightning：

I'm not saying Doom 2016 is not a fast-paced game.

I'm just saying that the base movement speed in Doom 2016 is far slower than in previous ID FPS titles.

aha... where is the shovel with which u can dig out the one who cares?
it fits very well to the gameplay and thats the reason why it feels like it does.
maybe ur just working for a competitive company? ea games maybe? ...or ubi?
otherwise this whole thread doesnt make sense :P
ur trying to compare miles per hour with kilometers per hour from two different universes with different physics