I'm a little late to the party, but I'd like to add a reply to some of the comments made by @OLADUSHEK.
I'll begin by attempting to draw some parallels with RM. First off, regarding the villagers issue. Like in RM, it is important to keep as many of your villagers alive as possible, particularly in early game. For RM, this means not losing villagers to boars, defending vs drush/trush, etc. For DM, this means not losing villagers to enemy scout or first ~5-10 units the enemy uses to raid, which in DM we refer to as "rushing". Of course it cannot be helped most of the time that you will lose a couple villagers; however, like in RM, the goal is to minimize your losses.
Second: strategies. I often hear phrases like "DM is just a spam fest" from players who are mostly unfamiliar in DM. I can understand where this viewpoint is coming from, but it is far from the truth. In RM, typically a player scouts around his/her base and keeps track of some simple details about the layout of the map (for example if enemy has forward golds) as well as what civilization both the player and the opponent is. Using this information, they choose from a small set of possible early strategies; for example drush, trush, scouts, etc. In DM, we begin by usually putting 2 vills to build a military building (barracks or stable) and 1 vill on house. We then send the scout to enemy base by using the minimap; this is all done in the first ~15 seconds. The key here is when sending the scout we take a quick view of the minimap in order to choose our build-order depending both on the layout of the map (if enemy has forward golds, where major hills lie, where the relics are, etc) as well as what civ the opponent is. To give a brief example, suppose I am playing a huns mirror on arabia and I notice that my opponent is very close to me with a hill directly between us. In this case, I may choose to do a 6-7-5 (6 stable, 7 barracks and 5 archery ranges) build order. In this case, I cannot simply full-queue every stable: it would cost 6750 gold to queue 6 stables with 15 pala each, and 3600 gold to queue 5 archery ranges with 15 HCA each, thus costing more than what my starting 10000 gold allows for. Instead, I would choose to queue 10 pala in each of the 6 stables, thus costing a total of 3600+4500 = 8100 gold. This gives me gold room to produce early rams (which are very effective at tanking fire once the paladins from both player are gone), make a couple monks to get relics, and finally leaves a small buffer to allow me to adjust what units I'm making depending on how the game is going in the first couple minutes. In this scenario I will also run out of food much earlier than by doing other popular build orders due to the heavy emphasis on both paladin and halbs - the most popular huns build order is the 4-6-6. From here I use my scout and first couple paladins to raid villagers while analyzing the build order of my opponent; if I feel they are going with a more passive strategy with less stables and more archery ranges/barracks, I'll send my army forward to take the hill very early. On the other hand if my opponent is making an even more aggressive early-game strat, I would choose to hold back and mass my army for an extra 30 seconds to 1 minute before moving out. Rather than write a massive essay about how strategies in DM evolve over the course of the game with consideration towards a plethora of factors, I will stop my explanation here.
The point I am trying to get across in the above writing is that DM has many of the same basic flavors as RM; the key difference is that they all happen at a much faster pace. When I first played aoc back when the game came out, I was an RM player just like the majority of new players; however, after giving DM a try, I felt that I could never go back to the slower pace of RM.
Lastly, for the points about DM having far less number of players. DM has always had a smaller player-base than RM, nobody can deny that. However, DM did have a much bigger following at different points in the past, even leading to DM being included in the famous WCL tournaments. In DE, I believe a major factor (amongst many other factors) that is limiting DM from being played more is simply the maps in matchmaking. I can't tell you how many DM players I have spoken to that have either switched to RM or quit Aoe2 entirely since DE has come out. For example, you would have a very difficult time finding DMers who actually enjoy playing on Arena, Mountain Pass, Hill Fort, Continental, Bog Islands, etc.
I hope this super long post is enlightening to anyone who might have some interest in DM. Lately, DM has seen a fair amount of growth, and I am still a strong believer that it deserves its relevant place within the broader aoe2 community.
I'll begin by attempting to draw some parallels with RM. First off, regarding the villagers issue. Like in RM, it is important to keep as many of your villagers alive as possible, particularly in early game. For RM, this means not losing villagers to boars, defending vs drush/trush, etc. For DM, this means not losing villagers to enemy scout or first ~5-10 units the enemy uses to raid, which in DM we refer to as "rushing". Of course it cannot be helped most of the time that you will lose a couple villagers; however, like in RM, the goal is to minimize your losses.
Second: strategies. I often hear phrases like "DM is just a spam fest" from players who are mostly unfamiliar in DM. I can understand where this viewpoint is coming from, but it is far from the truth. In RM, typically a player scouts around his/her base and keeps track of some simple details about the layout of the map (for example if enemy has forward golds) as well as what civilization both the player and the opponent is. Using this information, they choose from a small set of possible early strategies; for example drush, trush, scouts, etc. In DM, we begin by usually putting 2 vills to build a military building (barracks or stable) and 1 vill on house. We then send the scout to enemy base by using the minimap; this is all done in the first ~15 seconds. The key here is when sending the scout we take a quick view of the minimap in order to choose our build-order depending both on the layout of the map (if enemy has forward golds, where major hills lie, where the relics are, etc) as well as what civ the opponent is. To give a brief example, suppose I am playing a huns mirror on arabia and I notice that my opponent is very close to me with a hill directly between us. In this case, I may choose to do a 6-7-5 (6 stable, 7 barracks and 5 archery ranges) build order. In this case, I cannot simply full-queue every stable: it would cost 6750 gold to queue 6 stables with 15 pala each, and 3600 gold to queue 5 archery ranges with 15 HCA each, thus costing more than what my starting 10000 gold allows for. Instead, I would choose to queue 10 pala in each of the 6 stables, thus costing a total of 3600+4500 = 8100 gold. This gives me gold room to produce early rams (which are very effective at tanking fire once the paladins from both player are gone), make a couple monks to get relics, and finally leaves a small buffer to allow me to adjust what units I'm making depending on how the game is going in the first couple minutes. In this scenario I will also run out of food much earlier than by doing other popular build orders due to the heavy emphasis on both paladin and halbs - the most popular huns build order is the 4-6-6. From here I use my scout and first couple paladins to raid villagers while analyzing the build order of my opponent; if I feel they are going with a more passive strategy with less stables and more archery ranges/barracks, I'll send my army forward to take the hill very early. On the other hand if my opponent is making an even more aggressive early-game strat, I would choose to hold back and mass my army for an extra 30 seconds to 1 minute before moving out. Rather than write a massive essay about how strategies in DM evolve over the course of the game with consideration towards a plethora of factors, I will stop my explanation here.
The point I am trying to get across in the above writing is that DM has many of the same basic flavors as RM; the key difference is that they all happen at a much faster pace. When I first played aoc back when the game came out, I was an RM player just like the majority of new players; however, after giving DM a try, I felt that I could never go back to the slower pace of RM.
Lastly, for the points about DM having far less number of players. DM has always had a smaller player-base than RM, nobody can deny that. However, DM did have a much bigger following at different points in the past, even leading to DM being included in the famous WCL tournaments. In DE, I believe a major factor (amongst many other factors) that is limiting DM from being played more is simply the maps in matchmaking. I can't tell you how many DM players I have spoken to that have either switched to RM or quit Aoe2 entirely since DE has come out. For example, you would have a very difficult time finding DMers who actually enjoy playing on Arena, Mountain Pass, Hill Fort, Continental, Bog Islands, etc.
I hope this super long post is enlightening to anyone who might have some interest in DM. Lately, DM has seen a fair amount of growth, and I am still a strong believer that it deserves its relevant place within the broader aoe2 community.