Blood Swords
Ok, as a bladesmith, I would go with at least 1000 people.
During the smelting process, assuming you're also using a traditional furnace and melting the iron ore down through charcoal and such to pick up carbon, you're going to end up with a mix of steels. Some will be high carbon, and good for making a blade (or at the very least the edge) and a good amount of low carbon steel. The low carbon steel isn't good for a blade, as it won't get hard enough to hold an edge. However, you can use to form the core of a multi-bar construction sword, or the spine of a single-edged blade. Many katanas have a soft steel spine, since there was only enough good steel to use for the edge.
Next, you've got to work that steel bloom, folding over and over and over again to work out impurities it picked up in the smelting. This is going to burn off a fair amount of metal. Not a huge amount, but noticeable if you don't have any margin for error.
Once you get into the forging, you're going to lose more in the same way, through forge scale. Again, not a big deal if you've got plenty of material, but if you have a limited amount, you might end up making a dedicated kitchen slicer instead of a chopper.
After that comes grinding. You've got to even up the profile of the blade, and even if you've mostly hammered in your edge bevels, you've got to fully sharpen them. Plus, if you're making a sword, you want to make sure to work out the hammer marks and other issues that might cause for poor edge alignment or geometry.
By this point, you've lost 10-15% of the steel you started with when you pulled it out of the furnace. Perhaps more if you're unlucky. If you didn't forge your bevels and just ground them, that number could be up around 30%. Even more if you decide to grind in some fullers (NOT BLOOD GROOVES).
Now it's time to do your heat treat. You bring that blade up to a big past the critical point where the steel becomes non-magnetic, let it sit for a few moments, soaking and relaxing, then you take it out and let it cools slowly. This relaxes the grain structure and helps even out the carbon distribution in the steel. Do this at least a couple of times. Each time, you might loose some of the outer steel to scale or decarborizing as the hot metal reacts with the oxygen in the air.
Finally, you heat it up one more time. This time, instead of cooling slowly and letting it soften, you quench it, trying to bring it from about 1500F to 200F within a second or two. This snaps the crystal lattice into place, trapping the carbon atoms in between layers of iron atoms, so they disrupt the pattern and make it more difficult for the layers to shear along cleavage lines.
What do you quench it in? Depends on how hard you want it and what the composition of the steel is. Water can work ok for most steels, and better for some. Salt water allows for an even faster quench, and thus a harder blade, but it might be too aggressive and cause your blade to crack or even snap. Oil is good, but needs to be pre-heated to reduce viscosity and allow it to flow around the hot metal better so it can pull out the heat faster.
Now, this picture seems to be figuring the amount needed by weight. They might not have subtracted the weight of the hilt and pommel and other fittings, so you might save some material there.
"But, what about saving the work and just quenching in blood?" you ask. Sure, you need a lot less blood to simply plunge a hot blade into it and quench it. In theory, it might seem like a good idea, that maybe it could pick up some last minute carbon from the blood, since its an organic material. However, blood is indeed thicker than water. Just like the oil, it is viscous. Sadly, if you heated it enough so that it wouldn't just be a slimy insulator and slow the cooling of the steel, it would be too hot to bring the steel down to the level it needs to harden at all. How do I know? No, I haven't tried it. I've just read the paper of a metallurgist who did.
TL;DR: So, if you're going to forge a blade from the blood of your enemies, you want at least 1000 corpses' worth of blood. Preferably, you've had them as prisoners for a while and fed them a high-iron diet. You're going to lose a lot of material in the process, but it potentially possible. Just really not worth it.
During the smelting process, assuming you're also using a traditional furnace and melting the iron ore down through charcoal and such to pick up carbon, you're going to end up with a mix of steels. Some will be high carbon, and good for making a blade (or at the very least the edge) and a good amount of low carbon steel. The low carbon steel isn't good for a blade, as it won't get hard enough to hold an edge. However, you can use to form the core of a multi-bar construction sword, or the spine of a single-edged blade. Many katanas have a soft steel spine, since there was only enough good steel to use for the edge.
Next, you've got to work that steel bloom, folding over and over and over again to work out impurities it picked up in the smelting. This is going to burn off a fair amount of metal. Not a huge amount, but noticeable if you don't have any margin for error.
Once you get into the forging, you're going to lose more in the same way, through forge scale. Again, not a big deal if you've got plenty of material, but if you have a limited amount, you might end up making a dedicated kitchen slicer instead of a chopper.
After that comes grinding. You've got to even up the profile of the blade, and even if you've mostly hammered in your edge bevels, you've got to fully sharpen them. Plus, if you're making a sword, you want to make sure to work out the hammer marks and other issues that might cause for poor edge alignment or geometry.
By this point, you've lost 10-15% of the steel you started with when you pulled it out of the furnace. Perhaps more if you're unlucky. If you didn't forge your bevels and just ground them, that number could be up around 30%. Even more if you decide to grind in some fullers (NOT BLOOD GROOVES).
Now it's time to do your heat treat. You bring that blade up to a big past the critical point where the steel becomes non-magnetic, let it sit for a few moments, soaking and relaxing, then you take it out and let it cools slowly. This relaxes the grain structure and helps even out the carbon distribution in the steel. Do this at least a couple of times. Each time, you might loose some of the outer steel to scale or decarborizing as the hot metal reacts with the oxygen in the air.
Finally, you heat it up one more time. This time, instead of cooling slowly and letting it soften, you quench it, trying to bring it from about 1500F to 200F within a second or two. This snaps the crystal lattice into place, trapping the carbon atoms in between layers of iron atoms, so they disrupt the pattern and make it more difficult for the layers to shear along cleavage lines.
What do you quench it in? Depends on how hard you want it and what the composition of the steel is. Water can work ok for most steels, and better for some. Salt water allows for an even faster quench, and thus a harder blade, but it might be too aggressive and cause your blade to crack or even snap. Oil is good, but needs to be pre-heated to reduce viscosity and allow it to flow around the hot metal better so it can pull out the heat faster.
Now, this picture seems to be figuring the amount needed by weight. They might not have subtracted the weight of the hilt and pommel and other fittings, so you might save some material there.
"But, what about saving the work and just quenching in blood?" you ask. Sure, you need a lot less blood to simply plunge a hot blade into it and quench it. In theory, it might seem like a good idea, that maybe it could pick up some last minute carbon from the blood, since its an organic material. However, blood is indeed thicker than water. Just like the oil, it is viscous. Sadly, if you heated it enough so that it wouldn't just be a slimy insulator and slow the cooling of the steel, it would be too hot to bring the steel down to the level it needs to harden at all. How do I know? No, I haven't tried it. I've just read the paper of a metallurgist who did.
TL;DR: So, if you're going to forge a blade from the blood of your enemies, you want at least 1000 corpses' worth of blood. Preferably, you've had them as prisoners for a while and fed them a high-iron diet. You're going to lose a lot of material in the process, but it potentially possible. Just really not worth it.