commitandquit

Different

[commitandquit @commit_andquit]

Great video! Have you seen the recent news about severe drought across the US? Potential catalyst for the stock. At a glance it looks like the most impacted areas align with datacenter buildouts. It is my perception that the water usage is the biggest political pain point for datacenter builds. I think you have a the potential for a huge catalyst here. Move where the puck is going, fucking love it. I'm in.

[Aaron Chan @RecurveCapital]

$CCOI's deal today was below my estimated value, but still quite positive for the balance sheet and credit story. Based on feedback today, these 10 facilities had annualized costs of $7m/year.

If we run these proceeds through Cogent's playbook that I wrote about here (https://t.co/zOI7iRqVhv), leverage will declined by about 0.9x (lower debt and higher EBITDA from lower opex) and FCF will improve by at least $20m/year (lower opex + interest savings as they repurchase 6.5% notes). This should really help the equity going forward.

Next step is the consents/amendments from bondholders which has been in motion for many weeks already, hopefully it won't be too long until we see that come through. Once the refi of the 2027s is done, Cogent will have years of runway before its next refi. By then, it won't matter anyway - they will be on much firmer footing.

I am skeptical that 100% of the remaining facilities will sell (especially the ones <1 MW), but Fort Worth is the biggest one and probably the crown jewel of the whole portfolio to most buyers. I think an outcome in the $100-200 million range would be very positive, especially for EBITDA and FCF. Lower opex, more cash on the balance sheet.

I am sure Cogent was reluctant to enter other LOIs until this deal went definitive since any of these 10 could have been bundled with the remaining 14 to make other deals. I would expect some pretty quick movement on the most attractive remaining DCs.

I tried to map out what could happen as they make progress on all these fronts, starting from Q1 results and going through year-end, assuming they sell another $150M of DCs and get some on net growth driving EBITDA, plus additional synergies.

In the end, this is step 1 in a journey toward lower leverage levels, but I'm not particularly interested in a distressed equity becoming non-distressed (although those situations generate good returns). We need the growth story to kick in for it to be a great story going forward.

[commitandquit @commit_andquit]

$CHTR is being eaten alive by fixed wireless hence their push into mobile. Fixed wireless is effortless to install (no truck roll self-install), reliable due to lack of physical lines to the edge, and the customer service is a way better from the providers.

I also don't understand this vCMTS thesis. I suppose for the DOCSIS 4.0 upgrades and theoretical rack space reduction but there is no giant capex push to deploy them as far as I'm aware. Customers aren't switching due to lack of bandwidth.

As for the Edge AI thesis, bandwidth is not the constraint, it's latency, and the HFC plant has higher latency. Not to mention lead times for install vs a wireless product for a greenfield site.

I do find the hub site GPU deployments to be curious. Very good idea however deployments at these bloated cable companies take..a .. very .very .. long time. That plus hubs are not created equal.

idk man. I'm curious what you're seeing. CHTR is a money making machine but it's bloated to a point of no return IMHO.

[commitandquit @commit_andquit]

Both! Fiber connectivity has really taken a back seat in the datacenter hype even though it's a key element for site selection. These tier 1 internet backbone companies have been overlooked so far. Not to mention their access to existing infra. Many are actually installing GPUs in their hubs.

The balance sheet looks awful and the gamble here is their ability to monetize their assets, pay off the debt and stem the hemorrhaging from the Sprint legacy network; however, datacenters, fiber and IPv4 space is like selling water in a desert at the moment which is why I think there is so much upside.

[commitandquit @commit_andquit]

@MartinShkreli @csvoss @Etched @neurophos @OLIXComputing @taalas_inc Love the siphon attention strat here. I'm in. Like the thesis. You can believe in the tech, fund it, and make some money along the way.

[commitandquit @commit_andquit]

Dude, I wish I knew. I cant find anything. I bought it in anticipation of AI compute meme euphoria assuming retail would eventually find it (success). Has all the right buzz words and themes. I agree with @DrocDolf seems pretty scamy. He’s done a lot of research on it. The AI made podcast on their site doesn’t help their case. I exited.

[commitandquit @commit_andquit]

@DrocDolf @fundmyfund Thats the name of the game. Then you move on…

[commitandquit @commit_andquit]

@bubbleboi Which ones were the dumps? Am i missing something? Been pretty solid. This guy just jerks off about his intel play all day.

[commitandquit @commit_andquit]

@DeepValueBagger @amitisinvesting Brutal dood

[commitandquit @commit_andquit]

@aleabitoreddit I heard Elon Musk call out the transformer bottleneck on the Norges Bank podcast back in 2024. Great episode. Got me to buy GEV.

[commitandquit @commit_andquit]

@coinbureau Why are rehypothicated assets allowed to be used as collateral? This is obviously going to continue happening.

[commitandquit @commit_andquit]

@pepemoonboy You’re betting a regional war with economically critical geography can be TACO’d. Never mind Iran has been preparing for (maybe even ‘asking for’) this exact moment for four decades. Good luck with that shit.

[commitandquit @commit_andquit]

@sentdefender This real!? Shit that camera man was on point!

[commitandquit @commit_andquit]

@TRIGGERHAPPYV1 I know that guy, that’s Eprey Jeffstein

[commitandquit @commit_andquit]

@sentdefender I’m one to stay outside and film a tornado, but this is on another level.

[commitandquit @commit_andquit]

$HGRAF

[commitandquit @commit_andquit]

Dude, I feel insane talking to normies about where we’re at with AI. Personally I’m doing more complex projects and interdomain type work I otherwise would not be doing pre-AI. It's not just productivity that is increasing, it also allows lateral moves into different knowledge domains.

[commitandquit @commit_andquit]

@TheSqeakyMouse