Legion Map Comparison

[megasus]

Edit 06/30/17: Updated for Patch 7.3

Now that the Minimap of the Broken Isles has shown up, I've decided to make some measurements with it and show how it holds up
compared to other continents.
This is just for fun and a way to provide some scale, not an xpac1 vs. xpac2 kinda thing.

Also keep in mind that we're looking at an unfinished map, meaning that the shape of zones might change and the possibility of
other islands being added throughout content patches.

Other than that of course: Bigger doesn't always mean better.
Certain scenes require more work by the artists, others not so much.

Huge empty fields like this desert in Uldum or this place in Icecrown help increasing the size of a zone,

yet there is visually not much going on.

Especially since MoP there is a noticeable improvement in terms of 'Level of Detail', content density and general polish,
so all these things have to be taken into consideration.

Moving on.
Measuring the size of continents is a simple process. Rather than making weird ingame distance approximations, all calculations are
based on minimaps which are extracted directly from the client and then assembled. You'll end up with a big .png file.
Unlike Worldmaps, Minimaps are not hand-painted but generated using actual terrain data. Everytime Blizzard makes noticeable (even small) changes to the world, they'll generate and deploy new minimaps for affected zones.
The most important thing though: Minimaps all have the same scale. Meaning a 100px distance on one map equals the same distance ingame as a 100px distance on another map.

You can verify this yourself by traversing a zone while measuring the time it takes you to get from one point of your minimap to another. Use the image editing software of your choice to measure the distance in pixels between those two points of your minimap. Basic math will now get you the distance you've traveled in px per 1 second. Do the same thing on a different continent and you should get the same value, proving that in fact both minimaps have the same scale.

Since this is the case, once we get the pixel count of a given zone - representing the surface area - we can easily use this number for comparisons.
Now to get the pixel count of a zone we can use Photoshop or other image analytic tools. However first the minimap image needs to be croped so that we only measure the landmasses and not the ocean and other non-relevant areas. I've also cut out PvP-only zones like Ashran or Wintergrasp as I consider those the same categoty as battlegrounds.
Removing the entire ocean isn't entirely fair considering it's often also part of the player experience but we deal with it for now.

The cropping process doesn't need to be perfect - a tolerance of about 0.5% is pretty acceptable imo.

Beneath is the result of this process. It's a downsampled version since I'm not able to upload a 400 gigapixel image.
New race/class zones as well as patch zones were also factored in and marked in different colors. Although I intentionally left out Thal'Dranath since it's just a very raw shape atm and will probably have a different shape when it gets released.

[huth]

Something else that bears mentioning: Unlike previous expansions, every zone is also going to be a max level zone this time, so the total useable area at cap is significantly larger than before.

[iindigo]

Not surprised the Northrend takes the cake. That place is massive. How do all of these continents stack up against EK and Kalimdor?

I think empty areas add to the “realness” of a zone. The newer super-dense style of zone constantly remind me I’m playing a game because everything on the map has an obvious purpose. More distributed areas feel more “world like”.

[Chickat]

Id prefer it to be a bit bigger, but its plenty big tbh. As long as theres content its fine.

[Ansi]

Really interesting ─ thank you for doing this.

[Zhira]

Awesome. Thank you for sharing this, dude! (or dudette)

/ Z.

[megasus]

How do all of these continents stack up against EK and Kalimdor?

Might do some calculations at some point, I think EK and Kalimdor combined would be about 3 times the size of Northrend.

[schwarzkopf]

Using your 'algorithm' for which is bigger - please tell me which of these two is bigger :



Point is - something can appear to be smaller because there is actually more stuff to pack in.

The only reliable way to compare is to do an in game walk of a specific distance on map, then get a scale in yards.

[micwini]

Not fair to count all the patchzones for the past expansions.

[megasus]

Using your 'algorithm' for which is bigger - please tell me which of these two is bigger :

Point is - something can appear to be smaller because there is actually more stuff to pack in.

The only reliable way to compare is to do an in game walk of a specific distance on map, then get a scale in yards.

Read the op. All maps have the same scale and I've provided methods to verify this.

Not fair to count all the patchzones for the past expansions.

Well it's not like I pretend there won't be any for Legion. Just a thing to keep in mind.

[Deleted]

Size =/ more content. Northrend was big, but mainly because of the grotesk buildings. Take those out of it and it would probably be smaller than MoP.

Same thing with WoD big but a lot of desolate places with nothing interactive about them.

[schwarzkopf]

Read the op. All maps have the same scale and I've provided methods to verify this..

Well you are wrong, just verified in game.

Same distance on each of your maps (flying - straight line).
WoD: 106 seconds
MoP: 142 seconds

[megasus]

Well you are wrong, just verified in game.

Same distance on each of your maps (flying - straight line).
WoD: 106 seconds
MoP: 142 seconds

Cool you managed to do the first step. Read again.

[schwarzkopf]

Cool you managed to do the first step. Read again.

So - your maps are not the same scale then ?

That's the inconsistency - either your maps are not the same scale, or your assertion is false.

My above comment was based on your maps being the same scale - as you put a lot of effort into getting the pixel counts etc.

The only remaining conclusion was that the assertion you made about the comparisons of the maps was off.

[Imnick]

Minimaps do actually have the same scale schwarzkopf, else they wouldn't work in game.
Your character doesn't speed up and slow down on the minimap when moving between areas, they always move at the same speed, so the scale is also the same.

The game directly interprets your character's X/Y movement to scroll the minimap, which is one continuous image.
If the different maps were different sizes, there is no way this could work.
It's fine for the larger maps because they approximate a lot more and don't show as much detail. You can visibly see your icon moving faster or slower in differently scaled maps when you press M.

[megasus]

So - your maps are not the same scale then ?

That's the inconsistency - either your maps are not the same scale, or your assertion is false.

My above comment was based on your maps being the same scale - as you put a lot of effort into getting the pixel counts etc.

The only remaining conclusion was that the assertion you made about the comparisons of the maps was off.

First of all for the way how minimaps work, they need to have the exact same scale.
The minimap display ingame loads these map tiles and moves them according to your movement speed. The marker that shows the position of your character doesn't move. If these tiles were not based on the same scale they would need to have a scale factor associated with them and we would be able to find such a field in the client's dbc files.
However there is a way to validate that all minimaps have in fact the same scale:
First you need full-resolution versions of the minimaps of the zones you want to check.
Now position yourself ingame with a flying mount on one map and screenshot your current minimap location. Now fly as long as you want in a straight line.
When you stop, note the time and again screenshot your current minimap location.
Now open the full-res minimap file in a image editing tool and find the two minimap spots, measure the distance in pixels between them.
Let's assume it took you 100 seconds to traverse 200px. That would mean 2px/second.
Do the same thing on a different map.
Let's assume that took 150 seconds to traverse 300px -> 2px/second
Both values should be near identical, meaning that both maps have the same scale.

[melodramocracy]

I personally hope they continue making much smaller landmasses. After all, less is more, and who wants to actually travel anywhere?

Oh, and make the ground pretty flat. Being impeded by rocks n stuff is terrible.

- - - Updated - - -

First of all for the way how minimaps work, they need to have the exact same scale.

Here's how I've done it in the past:

-Pick a spot on any map, on a large water area, using a water walking mount or pot. You want to use water because elevation changes on the ground will alter the results.

-Take a screenshot of the full map.

-Get a stop watch, and run directly east (or north, whatever) for a set amount of time. Say... 45 seconds.

-When you come to a stop, take another screenshot.

-Repeat this process for every continent if you wish.

-load your screenshots into a photo editor, overlaying them all, and ensuring that your start & end points line up exactly over one another. Now you have a fairly descent comparison of 1 landmass to the next.

[megasus]

If you really want a somewhat reliable way to see actual map scale from one to the next, here's how you can do it:
[...]

I've provided a pretty solid way of validating that the scale is the same, I don't know what else there is to discuss about that.

[melodramocracy]

I don't know what else there is to discuss about that.

I was unaware that providing additional input wasn't allowed here.

[Paula Deen]

OP is correct, but that doesn't really mean anything. MoP had a tiny as hell main continent and it had arguably the most content.