Voluntary Payout Reductions — and How to Set One Up

[Grimm]

Or borderline gaslighting.

[littleskunk]

Well another unsolvable math problem. I guess we can just wait until he gets the core message. 2 people are claiming they believe the payout rate is too high and you still manage to find evidence for quantum mechanics in there.

[arrogantrabbit]

@Grimm

because this is not quantum mechanics, both can’t be true and false at the same time.

Lol , it’s not, but there is no contradiction

@Grimm

You say no SNO is ever going to make a profit, specifically “it’s never going to pay for your drive”. That is your statement

I never said that. I said it’s not meant to be profitable by design. I also said somewhere that “it was possible for a while”, and disk and electricity prices are not homogeneous across the world, so there will be some long tail of people that will always be able to make some profit regardless of payout rate. Imagine some raspberry pi with a small hdd running in the corner where electric rates are 0.10/kwh – I can easily see can be profitable.

Where I live, electricity is $.45/kwh. Just one my server at home cost me $60/month to keep online. I need it always online. I would have been paying that myself. Now Storj pays about $24 for that. I’m grateful. What’s not to like? If it was paying $10 – I would be also grateful.

So yes, profit is possible for some, but it’s not the design intent. The intent is “offset costs” and “monetize idle, otherwise wasted, already online capacity”.

It makes no sense for storj to rent VPS with block device to run nodes. And yet, some home operators manage to build hardware today.

BTW, in the bootstrap stage storj was paying over 10x It was unsustainable but served a purpose.

[alpharabbit]

@arrogantrabbit

It makes no sense for storj to rent VPS with block device to run nodes.

In my universe they do just this. They call it select.

[Vadim]

@arrogantrabbit

BTW, in the bootstrap stage storj was paying over 10x It was unsustainable but served a purpose.

It was a miracle, I builded up all my setup from this money.

[Grimm]

@arrogantrabbit

If you want storj to pay for your drive – this is not happening, by design. At some point it was possible accidentally, when drives were cheap, but that was not the intention.

I was paraphrasing but I am pretty sure it means the same dam thing.

@arrogantrabbit

I also said somewhere that “it was possible for a while”,

you did not said that exactly, you are paraphrasing, like I did!

So you correcting yourself by saying that it is possible on the long run with cheap hardware and cheap power, but it shouldn’t be?

[Grimm]

I don’t like being toyed with and I don’t like mind games.

And if you are the best storj has to offer I am worried.

So storj objective is to always be unprofitable to the SNO, if I send an email to the CEO he will 100% agree with this?

[littleskunk]

@Grimm

I don’t like being toyed with and I don’t like mind games.

You are trying to falsify other people so hard that you take everything out of context. We can’t help you there. You are the person that wants to play mind games. You see unsolvable math problems and quantum mechainics where there was a simple overlapping message. I get paid too much to the point that I would have to scale up. He is saying that is not what storj should be. I agree on that even if I silently hope we can meet somewhere in between. I do see a payout range that would still allow me to make a small profit and prevent me from scaling as agressive as I am considering right now.

[agente]

@Grimm

I am not asking you to send, show me where I can find a 20TB new drive for 270€.

I sold multiple 18TB at 350 right now (used 3 years). This is the real on the road price
(Europe market)

[Grimm]

@arrogantrabbit

If you want storj to pay for your drive – this is not happening, by design. At some point it was possible accidentally, when drives were cheap, but that was not the intention.

That was his initial statement

later he said

@arrogantrabbit

I also said somewhere that “it was possible for a while”, and disk and electricity prices are not homogeneous across the world, so there will be some long tail of people that will always be able to make some profit regardless of payout rate. Imagine some raspberry pi with a small hdd running in the corner where electric rates are 0.10/kwh – I can easily see can be profitable.

Still quite different from what you say, and I still think you are an outlier!

@littleskunk

I am currently running 8 drives with 100W total consumption. @ 34 Cent. → 300€ per year or 25€ per month.

8 times 20 TB drives with only 50% usage on them is about $120 per month or a little under $1500 per year. With egress traffic it gets $150 per month / $1800 per year.

@littleskunk

I just checked my last invoice. Early 2025 I paid 270€ for a 20 TB drive.

we already had this discussion although I will still say, 8x20TB drives half full in germany is something comendable.

@littleskunk

I do see a payout range that would still allow me to make a small profit and prevent me from scaling as agressive as I am considering right now.

You still say that you can buy a requalified/refurbished drive TODAY and still make a profit, agressively.

You can’t both be correct and wrong at the same time, maybe he is wrong, that is fine, maybe you are an outlier that we can’t make decisions upon, that is fine.

No I am not trying to falsify anyone, or taking anything out of context, I care about correctness and understanding what the hell is going on and what the plan is for SNO.
Feel free to apologize!

I already did when I was in the wrong with you, I will post it here if I have to, in summary I didn’t read your setup and I assumed you only had 4 drives and you where lying about your numbers, still 8 drives half full in Germany really nice, I still think it’s an outlier for an SNO, and profitability with new or used drives is tight at best.

You are a Storj employee, that is your function here!

[arrogantrabbit]

@Grimm

So you correcting yourself by saying that it is possible on the long run with cheap hardware and cheap power, but it shouldn’t be?

No. I’m saying that it is not meant to be possible. The intent that it’s not possible for majority. It does not mean it’s impossible for anyone.

I think you are thinking in absolutes. It’s a spectrum. It does not have to apply to everyone — just the majority. Of course there will be outliers.

[arrogantrabbit]

@Grimm

You still say that you can buy a requalified/refurbished drive TODAY and still make a profit, agressively.

Well, yes. That’s why payout is too high.

Look, I can buy $400 drive today, move my node there, and at $20/month it will pay for the drive in 20 months.

[littleskunk]

And for the record. I still agree to that. Yes my setup is violating that intend. I don’t have a problem with that. I would still agree and come up with recommendation to meet somewhere in the middle.

[Grimm]

@arrogantrabbit

I think you are thinking in absolutes. It’s a spectrum. It does not have to apply to everyone — just the majority. Of course there will be outliers.

I agree, and my point is and always was, we can’t make policy or business decisions at this scale based on outliers.

@arrogantrabbit

Look, I can buy $400 drive today, move my node there, and at $20/month it will pay for the drive in 20 months.

ok, there is a bit to unpack here, I disagree, power is most European countries is expensive and we pay VAT, some pay upwards of 23%, on top of the power or buying a hard drive if you buy it from a business.

So it will not be 20 months you also have to fill up that drive, I have a hard time getting 10TB and I have been at this for 4 years, has so a lot of EU SNOs.

I am lucky if I get $11 per month and I have Fiber Gigabit, maybe I am the opposite outlier but for a single IP it is what it is for me.
I think the normal for EU is about 12 to 18TB per IP address maybe $15 a month.

@arrogantrabbit

Well, yes. That’s why payout is too high.

If someone is a good little SNO doing what is recommended either by alexey or some other official (dont do multiple ip’s, single dedicated drive, etc), any european SNO is hardly going to make a profit on $15 a month, specially on a fresh start or for some reason the drive dies, adding drives or upgrading is unfeasable, convertion to fiat and what not.

Now if someone runs multiple IPs, that’s another story, that is the outlier.

If you want to reduce the payment, the outlier has to be the norm agree? So you have to drop the /24 restriction.

Or you don’t do anything at all because we do not know how many outliers are if there are that many and how much data they concentrate, so it is the best interest of Storj to just keep SNOs has it is, you don’t want an exodus that might break something or concentrate too much of something correct?

So the little “dumb” SNO has to be a part of it and even him has to make some profit.

And lets face it, if the recommendations of single drive nodes and what not is to be taken seriously, this has to have some sort of profit, why would I buy a drive to loose money?

There has to be some motivation for SNOs, doing it for free while others make a buck not a good motivation, why don’t people go drive for Uber out of their own pocket…

Storj can dial however it wants to adjust the “market”, i can agree with that, but that is a very finicky dial, the conditions are not the same has before, everything is more expensive, and you have outliers, one wrong move and it’s chaos.

I understand what you mean by to many nodes, half of the pool is free space, but again one wrong move and we might be doing emergency repairs or worse, that’s reputation down the drain on an already shaky ground, bad for business.

[Grimm]

A fine example from a Storj employee , congratz

So it is now the official recommendation from Storj to run has many nodes has possible on a single IP address on the same location so you can cut the payout?

[arrogantrabbit]

@Grimm

12 to 18TB per IP address maybe $15 a month

Hmm. This does not compute. 12TBM is 12*.15 = 18. And this is storage only. Egress is extra. This is my 12TB node. The 4-year old one. The largest one. it pays $20.

Fun fact – note how repair traffic is double the useful egress… That’s the cost of node turnover when operators who are running on razor margins suddenly don’t make profit and leave.

@Grimm

disagree, power is most European countries is expensive and

So you want me to disprove my own point?

Ok, so it won’t be 20 month, it will be 40. 5W translates into 3.6kWh/month (units gore xkcd: Uncanceled Units). So even if you pay 50 cents – it’s like couple of euro. But you should not be using drive exclusive for in the first place…

@Grimm

I am lucky if I get $11 per month and I have Fiber Gigabit, maybe I am the opposite outlier but for a single IP it is what it is for me.

Again, explain. Show the screenshot. I don’t understand how do you get $11 with $12 TB stored, at $1.5/TBm.

@Grimm

specially on a fresh start or for some reason the drive dies, adding drives or upgrading is unfeasable, convertion to fiat and what not.

Another reason not to run naked drives for storj. My nodes are on a separate dataset on the single pool where everything is stored. Pool is redundant. It won’t fail.

@Grimm

If you want to reduce the payment, the outlier has to be the norm agree? So you have to drop the /24 restriction.

I don’t know what’s an outlier at this point. I don’t understand how are people hosting 30012 does under the same roof with storj being oK with this. My earlier sentiment:

@arrogantrabbit

No, the question is how is such concentration is tolerated by storj.

If I run two nodes on my single host I’m the villain. But if I run 2892 nodes in the same shed — I’m valued member of node operator community. Go figure.

So no, I’m running one node per location (except one place, where I run two, on the same IP) and that’s why I have 8 of them and not 80.

@Grimm

Or you don’t do anything at all because we do not know how many outliers are if there are that many and how much data they concentrate, so it is the best interest of Storj to just keep SNOs has it is, you don’t want an exodus that might break something or concentrate too much of something correct?

Yes, except even when looking at it as at a black box – we see there are plenty of nodes and nobody leaves. Meaning, payment can be reduced.

@Grimm

And lets face it, if the recommendations of single drive nodes and what not is to be taken seriously, this has to have some sort of profit, why would I buy a drive to loose money?

You won’t, and that’s the precisely the intended effect!

@Grimm

I understand what you mean by to many nodes, half of the pool is free space, but again one wrong move and we might be doing emergency repairs or worse, that’s reputation down the drain on an already shaky ground, bad for business.

I don’t follow the logic. How does amount of free space relates to the durability?

[littleskunk]

@Grimm

A fine example from a Storj employee , congratz

Thanks I guess. But there are not a lot of employees left. Not really a challenge these days…

@Grimm

So it is now the official recommendation from Storj to run has many nodes has possible on a single IP address on the same location so you can cut the payout?

Oh no you don’t want to go down that rabbit hole. I will give you 3 different answer depending on which part of me you are asking. As an employee the answer is no. We want high quality nodes. As a competing SNO I will say yes go for it. Run as many nodes as you can. Lets see how good your success rate is going to be. And the third anwer would be to share knowledge about the most efficient way to do it. From one friendly SNO to another.

[Grimm]

@littleskunk

Oh no you don’t want to go down that rabbit hole.

I do!
You represent Storj like it or not, you have to set the example!
Whatever you do or say here represents Storj.

[littleskunk]

@Grimm

You represent Storj like it or not, you have to set the example!

You might have a wrong understanding of my role. I am a QA engineer. I am getting paid for finding bugs. I will break things all day every day. I have done much worse. In the early days it was possible to self sign a 2PB order and hand that in for payout. I did that in production to also proof that our checks for finding cheaters are not working. This is my job. I am getting paid for playing the bad node. In return I will create bug reports for any loop hole I find so that storj has a chance to fix it.

@Grimm

Whatever you do or say here represents Storj.

So thats all? I can continue running my nodes the way I like as long as I don’t talk about it? How is that going to change the result?

[Grimm]

@arrogantrabbit

Again, explain. Show the screenshot. I don’t understand how do you get $11 with $12 TB stored, at $1.5/TBm.

I said I barely get 10TB, maybe I didn’t explained myself correctly.
the 12 to 18 was my estimate from other EU SNO reporting.

but here is my screen shot.

estimate for this month, today was an outlier, got 100GB or egress.

here is another one, probably missing a few bucks due to the token losing some value recently

@arrogantrabbit

Yes, except even when looking at it as at a black box – we see there are plenty of nodes and nobody leaves. Meaning, payment can be reduced.

I rather not poke that bear, luckily for me it’s not my call.

@arrogantrabbit

You won’t, and that’s the precisely the intended effect!

You honestly think that the network will survive and the business will be ok if there is no motivation to keep drives running in Storj?
I don’t but maybe that is just me.

@arrogantrabbit

I don’t follow the logic. How does amount of free space relates to the durability?

Free space correlates with your claim that we have to many nodes, it can be correlated that way, but those nodes with free space also have data, if too many of them go away too fast it’s an issue.