Mark I CASUC 1310EU/t

Metalchic · Feb 26th 2012

i'm assuming thats correct (title) becasue the reactor planner says Mark I-I-SUC EA and the ice will be added continualy at all times, even when theres an overflow kinda.

anyway i built this reactor and it relies on RP, EE and IC2 for its operation. technicly it imputs 5.7 ice blocks per second but the reactor planner doesnt support imputing decimals. anyway its total size is 4x4x2 blocks and according to the reactor planner it outputs 1310 eu/t (my question is on this cus i cant measure it) and i need help with the transmission of the energy.

the problem seems to be with the HV transformer. i used a setup before where i had to much EU for glass fibre cables but not enough for the most effiecient via 4xins HV cable where i used an HV transformer with redstone to transmit the 2048 packets with the minimal energyloss. but idk if they changed it or i'm doing it wrong or what becasue when i hook the redstoned transformer up to the 1300eu (trying to transform it to 2048 packets) it vapotizes. as you can see in my pic i have 6 ganged reactors but the one on the far end away from the MFSU aray is only producing 1000EU/t~ due ot cable losses. so did they change it so i cant turn the 1300EU packets into 2048 packets without some convoluted mess of transformers or am i doing it wrong?

pwnedgod · Feb 26th 2012

Using BCIC2Crossover will make the reactor chambers work with pipes.

Metalchic · Feb 26th 2012

Quote from pwnedgod

Using BCIC2Crossover will make the reactor chambers work with pipes.

excuse me? i think you missed the problem. the reactors are fed directly with RP2 Transposers and a timer set to 0.350 seconds each, the ice comes out of 2 Energy Condensers fueled by a single MK3 Collector. adding chambers (and making the transposers incompadible) would increase the complexity and size significantly, technicly removing a reactor chamber can reduce the size to 4x4x1 and reducing its energy output to i think 800EU/t~ i havent done the research or testing.

the problem i'm having is with the transmition of the EUs not the generation.

edit: i feel i sounded rude

AkhkharuXul · Feb 26th 2012

Maybe I'm missing the point, but I'm not sure why you want to up-convert it to 2048 EV packets when you could just use a HVT to down-convert it to 512 packets and run the glass fiber cable you wanted to.

- I would just use one 4x insulated HV cable going into the high side of a HVT then glass fiber out (for each reactor) [just make sure the HVT doesn't get a redstone signal], assuming the glass fiber cable isn't too long you'll have zero loss that way.

You could hook multiple reactors up to a HVT (to use less of them) but I'm not sure what would happen if you send it more then 2048 EU / tick (which would happen since you said each reactor is ~1300 EU/t).

The pic is rather confusing as I can't see where/how you were trying to hook up a redstoned HVT that was failing. Also I don't like the wall of cables on the side, as running cables like that causes rampant cable splitting calculations to happen (painters are your friend).

Metalchic · Feb 26th 2012

Quote from AkhkharuXul

Maybe I'm missing the point, but I'm not sure why you want to up-convert it to 2048 EV packets when you could just use a HVT to down-convert it to 512 packets and run the glass fiber cable you wanted to.

- I would just use one 4x insulated HV cable going into the high side of a HVT then glass fiber out (for each reactor) [just make sure the HVT doesn't get a redstone signal], assuming the glass fiber cable isn't too long you'll have zero loss that way.

You could hook multiple reactors up to a HVT (to use less of them) but I'm not sure what would happen if you send it more then 2048 EU / tick (which would happen since you said each reactor is ~1300 EU/t).

The pic is rather confusing as I can't see where/how you were trying to hook up a redstoned HVT that was failing. Also I don't like the wall of cables on the side, as running cables like that causes rampant cable splitting calculations to happen (painters are your friend).

Display More

i suppose i hadent thought about downconverting for transfer, that would eliminate the need for transformers at the other end i suppose. the wall of cables was me attempting to get more transformer capacity than i had output (otherwise there would be bottlenecks) and i didnt feel like pulling out a calculator to figure it out.

how i had origianly planned it was that each generator would have a transformer attached (the cables are coming out of the side of the closest chamber not the end) sideways and that would leave enough room on the edge of the dirt for a redstone torch or lever. attaching the 3dot side of the transformer to the reactor, upon appling redstone the transformer vaporizes. but if i were downconverting to 512packets for glass fibre cables i would only need 1 msfu instaid of a gang of transformers to collect the energy, (technicly i woudlent use 1 msfu but whatever).

also i wanted to point out that i'm trying to run a single mascerator with the theoretical max of 20 overclocker upgrades (thats why i made this)

tszaboo · Feb 26th 2012

MFSU only accepts 512 EU/t. If you want to store that energy, you need at least 3 per reactor.

xenoflot · Feb 26th 2012

> also i wanted to point
out that i'm trying to run a single mascerator with the theoretical max
of 20 overclocker upgrades (thats why i made this)

20 OCs want 3325 EU per operation. You need at least 1 Energy Storage Upgrade since the Macerator's internal storage is only 625 EU.
You can supply the 3325/tick a couple of ways. Have a bank of 7 MFSU supplying 512EU/t power over a single glass fibre cable.

In theory you could have a couple of backward HV transformers between the MFSU and the Macerator but I'm not sure if they'll support 3 transformer upgrades and run off EV. That needs testing Still, that's not really a practical suggestion. You'd still need a bunch of MFSU to feed the transformer to make EV.

Metalchic · Feb 27th 2012

Quote from xenoflot

> also i wanted to point
out that i'm trying to run a single mascerator with the theoretical max
of 20 overclocker upgrades (thats why i made this)

20 OCs want 3325 EU per operation. You need at least 1 Energy Storage Upgrade since the Macerator's internal storage is only 625 EU.
You can supply the 3325/tick a couple of ways. Have a bank of 7 MFSU supplying 512EU/t power over a single glass fibre cable.

In theory you could have a couple of backward HV transformers between the MFSU and the Macerator but I'm not sure if they'll support 3 transformer upgrades and run off EV. That needs testing Still, that's not really a practical suggestion. You'd still need a bunch of MFSU to feed the transformer to make EV.

Well i have the gang of reactors feeding a gang of 18 MFSUs that then feed into the maserator, there currently all full (takes about 3 minecraft days for the reactor ro fill them all up,) equaling about 9216EU/t, i'll backup my world and try out the 3-4 transformer upgrades idea.

what happens is that it will procsess a half stack (32 ore into 64 dust) in a fast speed i havent calculated or clocked it but with 64 storage upgrades it will deplete its internal storage buffer seemingly much faster than i can put EU into it.

edit: breakthrough! i threw a whole stack of transformer upgrades into the mascerator and it didnt deplete its internal buffer! i hooked up a redstoned HV transformer (HV>EV) but it doesnt have the capacity (2048EU/t max) by itself to run the maserator. i'm guessing that each transformer upgrade gives 4 times the EU/t pull potential of the previous number of transformer upgrades i'm testing that now. what worries me is that if i have a machine with 3 and it needs 4 to take EV it might blow up with the force of a mass fab.

(if my math is correct on the 4 times thing a machine with 64 transformer upgrades has a max capacity of 590295810358705651712 EU/t

edit2:after moar testing, i was able to get 18 OC upgrades running with 16 Storage Upgrades and 3 transformer upgrades, with 20 OC upgrades you gotta have all 64 storage upgrades and theres no effect past 6 transformer upgrades for this ammount of imput EU/t, meaning that i'm not putting in enough and it takes about twice as long for the internal buffer to replenish than it takes to process the halfstack.

xenoflot · Feb 27th 2012

SCIENCE!

Consider also that the wiki states there's a hard limit of "about 8000EU/t" so my maths say that although you may be able to power more than 20 OCs, they won't provide any benefit.
Kind of like paying $10,000 for a kebab. The man will happily take your money but it doesn't get you any more kebab than you would've gotten for $7.

Metalchic · Feb 27th 2012

Quote from xenoflot

SCIENCE!

Consider also that the wiki states there's a hard limit of "about 8000EU/t" so my maths say that although you may be able to power more than 20 OCs, they won't provide any benefit.
Kind of like paying $10,000 for a kebab. The man will happily take your money but it doesn't get you any more kebab than you would've gotten for $7.

ITS ABOUT THE WHY NOT! NOT THE WHY!

anyway i found the breakpoint to be at 13 for the furnace i'll check if any of the others are differnt this puts the output at afraction over 1 tick per operation, i doubt it'd be noticable.

AkhkharuXul · Feb 28th 2012

Quote from Metalchic

ITS ABOUT THE WHY NOT! NOT THE WHY!

anyway i found the breakpoint to be at 13 for the furnace i'll check if any of the others are differnt this puts the output at afraction over 1 tick per operation, i doubt it'd be noticable.

Yeah, I noticed some discrepancies with the overclocker section of the wiki last night so I took a look then did some maths.
Apparently only 16 overclockers are needed to max a Macerator [3689 EU/t], 13 for a furnace [1351 EU/t] and 10 for a recycler [109 EU/t]. Anything after that and you're just paying more for the same kebab, also the "hard limit" is actually 32, since anything over 32 would be considered 32. (though things start getting out of hand past 18 overclockers)

xenoflot · Feb 28th 2012

Quote from AkhkharuXul

Yeah, I noticed some discrepancies with the overclocker section of the wiki last night so I took a look then did some maths.
Apparently only 16 overclockers are needed to max a Macerator [3689 EU/t], 13 for a furnace [1351 EU/t] and 10 for a recycler [109 EU/t]. Anything after that and you're just paying more for the same kebab, also the "hard limit" is actually 32, since anything over 32 would be considered 32. (though things start getting out of hand past 18 overclockers)

I wrote that section of the wiki. Maths is not my best subject so if you have some corrections, please explain how you got to them and I'll update it.

AkhkharuXul · Feb 28th 2012

Quote from xenoflot

I wrote that section of the wiki. Maths is not my best subject so if you have some corrections, please explain how you got to them and I'll update it.

I peeked at the code. The formula at the top for power consumption changed going from IC 1.64 to 1.70. Now it's 1.6^x instead of 1.5^x (where x is the number of overclockers) the speed increase hasn't changed though, it's 0.7^x. [so each overclocker makes the machine 30% faster and uses 60% more energy/tick.]

I did make one assumption that the energy_consume * operation_length = cost for 1 operation. [ Which would mean a standard machine (like the macerator) without any overclockers would process 1 item in 20 seconds, using 800 EU to do so. I didn't bother actually testing that though. ]

To figure out the maximum number of effective overclockers I solved for 1 since anything less then 1 tick gets increased to 1 tick (or else at a certain point the machines wouldn't actually consume any energy). So 400*0.7^x = 1 .... 0.7^x = 1/400 {0.0025} which is where I got 16 from (or 17, wasn't sure at first, then looked at how java handles integers and saw that 16 should be correct).

I made a lua script to do all the maths for me (cos I wanted to know how overclockers affected the operation costs): --- Warning, spoilers ---

Display Spoiler

xenoflot · Feb 28th 2012

From the original wiki text, it increases the EU/Op (cost) by X% and reduces the Time/Op (time) by X%. It further suggests that there is a hard limit of 1 op/t. I can't tell from the wording if the listed 8000EU/t is also a limit or a consequence of the 1 op/t.

I was basing my rules of thumb on the listed 8000EU/t limit which in retrospect may have been a poor choice. I should calculate it to the point where an operation takes only a single tick.

So we need to establish how long an operation currently takes. Then we stack the OCs and the point where the op drops below 1 tick is the limit. Of course this is tricky since different materials take different lengths of time to process.

Oy vey. Time to get the stopwatch and do a little testing.

AkhkharuXul · Feb 28th 2012

Quote from xenoflot

From the original wiki text, it increases the EU/Op (cost) by X% and reduces the Time/Op (time) by X%. It further suggests that there is a hard limit of 1 op/t. I can't tell from the wording if the listed 8000EU/t is also a limit or a consequence of the 1 op/t.

I was basing my rules of thumb on the listed 8000EU/t limit which in retrospect may have been a poor choice. I should calculate it to the point where an operation takes only a single tick.

So we need to establish how long an operation currently takes. Then we stack the OCs and the point where the op drops below 1 tick is the limit. Of course this is tricky since different materials take different lengths of time to process.

Oy vey. Time to get the stopwatch and do a little testing.

Display More

Ugh, do different materials take different lengths to process? (I thought process time was based on machine so .. yeah that might be my bad, I'll look)

Ed: ... I'm pretty sure it's based off machine type, not material type...

xenoflot · Feb 28th 2012

OK, did some testing. Your number of 20 secs per operation for a macerator is correct. Tested with both smoothstone and coal. I could've sworn that coal took longer to macerate but it appears not.

So Macerator and Extractor are 400t/op, Furnace is 128t/op and Recycler is 45t/op.

I define Maximum Benefit as the point where the length of a single operation first drops below 1 tick and is thus effectively 1 tick. Of course you're "wasting a small amount of power" but if you don't cross the threshold, you're not quite as fast as you can be.

Revised numbers using the 160% power in 1.71 are as follows:

Macerator: 17 OCs - 5903 EU/t
Extractor: 17 OCs - 5903 EU/t
Furnace: 14 OCs - 2161 EU/t
Recycler: 13 OCs - 176 EU/t

I shall now update my addition to the wiki. Thanks for helping out!
And sincere apologies for metalchic for highjacking her thread but it was in the name of science!

AkhkharuXul · Feb 28th 2012

Quote from xenoflot

OK, did some testing. Your number of 20 secs per operation for a macerator is correct. Tested with both smoothstone and coal. I could've sworn that coal took longer to macerate but it appears not.

So Macerator and Extractor are 400t/op, Furnace is 128t/op and Recycler is 45t/op.

I define Maximum Benefit as the point where the length of a single operation first drops below 1 tick and is thus effectively 1 tick. Of course you're "wasting a small amount of power" but if you don't cross the threshold, you're not quite as fast as you can be.

Revised numbers using the 160% power in 1.71 are as follows:

Macerator: 17 OCs - 5903 EU/t
Extractor: 17 OCs - 5903 EU/t
Furnace: 14 OCs - 2161 EU/t
Recycler: 13 OCs - 176 EU/t

I shall now update my addition to the wiki. Thanks for helping out!
And sincere apologies for metalchic for highjacking her thread but it was in the name of science!

Display More

I'm pretty sure due to rounding you need 1 less... and I'm not sure on the recycler cos you should have gotten 11 not 13 [and 11 would be 176 EU/t --- well 175, rounding again (13 would have been like 450)]

Then again I could easily do some testing of my own and just check the time for x items with 16 or 17 overclockers in a standard machine, and 13 or 14 in a furnace, as well as 10/11 in a recycler. [Should be pretty simple to see if they're equal since if it were bring it up to 2 ticks / operation instead of 1 it'd take twice as long ] ---- guess it's not that simple since I need to be able to supply 6k EU/tick to test standard machines.... *ponders TMI*

xenoflot · Feb 28th 2012

k, updated the wiki.

Not sure what you mean by the rounding.
Recycler - OC's, EU/t, number of ticks

10
110.0
1.3

11
175.9
0.9

I'd think that the game would look at 11 or more OCs taking 0.9 or less ticks, round up to 1 whole tick and thus 11 OCs becomes your effective limit.
Do you disagree?

AkhkharuXul · Feb 28th 2012

Quote from xenoflot

k, updated the wiki.

Not sure what you mean by the rounding.
Recycler - OC's, EU/t, number of ticks

10
110.0
1.3

11
175.9
0.9

I'd think that the game would look at 11 or more OCs taking 0.9 or less ticks, round up to 1 whole tick and thus 11 OCs becomes your effective limit.
Do you disagree?

Display More

1) You said 13 instead of 11 for Reclcyer OCs [hence my confusion].
2) Well you're correct, 0.9 gets rounded up to 1, however, 1.3 gets rounded down to 1 (so only 10 would be required, theoretically). [So, yes 11 will definitely make it 1 tick / operation, but 10 should be enough.]

3) The important fact is that -if- it's 16 instead of 17 OverClockers for the Macerator to run at 1 tick / operation then the difference in EU/t required is over 2k different [or 8 MFSU instead of 12 MFSU to power the thing]

-- I'll just install TMI to test

xenoflot · Feb 28th 2012

OK, just tested and it seems you're correct. The difference between 15 and 16 OCs over a stack of cobble was noticeable but I couldn't determine a difference between 16 and 17. Even a 0.3 tick per operation when multiplied 64 times becomes almost a second which should've been noticeable even with my Mk I Optical Measuring Device. We can thus conclude that the 1.3 is rounded down to 1 which adjusts my numbers by 1, as you say.

Thanks again for your feedback.

And to any of you who haven't been to university, this is what's called Peer Review and it's a foundation of modern science. SCIENCE ROCKS!

Share