« first day (2935 days earlier)      last day (2085 days later) » 

12:01 AM
I really like the design, and it uses Kaihua's well-received low-profile switches.
 
12:37 AM
good morning root access
 
Dog
12:49 AM
@allquixotic No, the only naswer is to tkae more. Until I'm immnune to deir effextz!
2
Speaking of which, -$1700 cumulative losses on bad trades now.
BUT 3 profitable trade position closures in a row! Only $6-$50 each time, but hey, it's becoming more consistent
I only need 60 more non-fuckups in a row to break even again
Also I somehow keep losing my Renko charts
 
1:07 AM
(that belongs on Matrix)
 
1:41 AM
....
they suddenly fixed the mobile connection at work
happy happy days
 
Any opinions on that Kickstarter?
 
@bwDraco just my usual
 
I mean, it's shot far past the goal, and I think the risk is low, but any more details?
 
2:02 AM
well, I've never used kalih or low profile switches before
 
my only opinion on Kickstarters is to assume every one of them is fake until proven otherwise with a working product in production
I've been disappointed far more often than not with them
 
Also... opinions on the upcoming GeForce RTX cards, and the state of competition in the GPU market? I can't shake the feeling that AMD is going to be locked out of the market completely until they have 7nm Vega available to consumers...
I'm skipping this generation. I cannot justify replacing an $850 graphics card so soon.
(EVGA GeForce GTX 1080 Ti FTW3 Elite Gaming)
 
@bwDraco I think them being 7nm or not is unrelated to their inability to keep up - it seems their architecture and/or drivers and/or aspirations for building a proper scale chip with a competitive number of shaders and/or GlobalFoundries' process is just not as advanced as TSMC, without regard for the fab size
 
All these are being fabricated on TSMC's 12FFN process designed specifically for NVIDIA. AMD may well gain a process advantage over NVIDIA by 2019, but I'm not sure if it's enough to close the gap.
AMD's GPU architecture just needs work. It's awfully inefficient especially at high frequencies.
 
they are still (AFAICT) on bulk fab instead of SoI, for instance, and they've for generations (when they even bother to compete with Nvidia for the top end) been around 10-20% slower
they haven't been ahead since they won the race to the first 28nm product with Evergreen (HD5000 series) and Nv was stuck preparing their counter for about 8 months
 
2:16 AM
They have a bunch of advanced compute capabilities like Rapid Packed Math for fast FP16 compute, but not many games are using them.
(Wolfenstein II, which was optimized specifically for AMD processors and GPUs, supports Rapid Packed Math and other Vega-specific technologies, but these games are few and far apart.)
Vega (the architecture) is great at lower frequencies, making it a superb performer when embedded in the likes of Raven Ridge (a fully-enabled Raven processor has 11 CUs or 704 SPs), but it doesn't scale as well as NVIDIA's best.
@allquixotic It's primarily architectural. Vega just does not scale that well.
It's almost as efficient as Pascal at low frequencies, but the architecture runs into serious power problems past about 1.2 GHz.
 
funny; Intel's going to have the same challenges when they plan to release a dGPU in 2020 since GMA is fine for small chips but will likely be crap as a dGPU
they'll have to totally redesign it
or maybe AMD will sell RTG to Intel lololololol
 
Vega 64 can match or outperform the GeForce GTX 1080 in many games, but it's a long distance away from the 1080 Ti and it uses about 50-60% more power.
@allquixotic This would not be a surprising move. Intel has been leaning on AMD Radeon graphics these last few months.
Look at the new 10nm NUC.
Intel's 10nm process is still in bad shape and OEMs aren't buying them, so they're selling them as part of a new NUC with a Radeon 540 as integrated graphics (8 of 10 CUs in Polaris 12).
Not to mention Kaby Lake-G, which is used in a few high-end laptops as well as a special high-performance NUC (Hades Canyon). Up to 24 CUs, with total (CPU+GPU+HBM2) TDPs of 65W and 100W.
 
2:34 AM
Vega M / Kaby-G is an awesome idea but they need GH model (more CUs!) in laptops, not just fucking NUCs
I mean if you can squeeze a GTX 1080 Max-Q in relatively thin and light laptops, and a 1070 Max-Q in the Razer Blade 15, how friggen hard would it be to cool a Vega M GH 100W in the same form factor?!
pretty sure the 1070 Max-Q uses as much, or more power than the Vega M GH
 
The 2018 15" models of the HP ENVY and Spectre x360 laptops are available with Kaby Lake-G.
 
@bwDraco yeah, but they have Vega M GL, which sucks
GL is roughly a 1050 or maybe 1050 Ti; GH is more solidly competitive with the 1060
 
This would need a full-thickness chassis. Not quite extra-thickness (e.g. high-end gaming laptop), but it would still be a lot heavier.
Fortunately, thin gaming laptops are getting more popular; I'm pretty sure this is doable with a ~20mm thickness.
(the 2018 HP ENVY x360 13, which is only available with Raven Ridge at 15W, will throttle under sustained heavy loads even if I put my Noctua NF-A12x25 beneath the intake and give it plenty of airflow)
> GTX 1080 Max-Q
This requires a near-full-thickness 15" chassis. Designs using this GPU typically hover around 18-19mm.
 
Dog
@allquixotic Why u so :( ?
 
(by today's standards, ~22-25mm is full thickness)
 
Dog
2:43 AM
Also I can see why standard stock markets have fixed opening hours (and sometimes fixed lunch breaks)... watching the charts so you don't lose out while having a "break" is a 24 hour job
 
@Dog oh, you know, the usual... mostly loneliness and related feelings
 
Dog
@allquixotic Avatar reflecting IRL sadness? :-(
 
@Dog yeah, but I also like it for unrelated reasons as it's the BSOD for a Windows Insider build
I would probably keep it, even if I weren't explicitly sad
 
(AIUI the Max-Q Design methodology imposes certain restrictions; NVIDIA requires laptops with Max-Q GPUs to have load noise levels no more than 40 dBA, weight no more than 5 lbs, and thickness no more than 20mm)
And I believe OEMs are not allowed to use Max-Q GPU variants if these limits are exceeded, and must use the standard mobile versions of these GPUs in larger systems.
 
Well, there's another thing - basically the bitcoin market's not working that well for nvidia - its traditionally been something that AMD/RTG has been a lot better
@allquixotic :(
 
2:52 AM
@JourneymanGeek AMD's GPUs are more compute-focused, which results in better performance for blockchain workloads.
 
@bwDraco Yup
and heh, I think everything is lining up for a late 2018/early 2019 build
(nand oversupply so dropping ram prices I hope, actual competition on the desktop side so tastier options, and actual, new and interesting video card options, though I have no clue if I actually want a RTX 2080, and if I can do multimonitor alongside my 980)
 
Well... I was wondering about using the RTX 2080 Ti alongside my existing GTX 1080 Ti.
Relegate the GTX 1080 Ti to PhysX acceleration.
But a heterogeneous setup like this is really unusual.
(not to mention that cooling will start to become a problem if both cards are at or near full load and an additional chassis fan will be necessary)
Also, a bit of a tidbit from the rumor mill (a very strong rumor, but take with a grain of salt as usual): NVIDIA is using NVLink for SLI.
Datacenter-grade GPU interconnectivity is reaching consumers.
 
its mentioned in the early stuff actually
@bwDraco the funny thing is my existing system can probably handle it
also, windows will do multi gpu with dx 12
 
DirectX 12 is not yet widely adopted, let alone multi-GPU capability.
 
(also, I consider the non TI x80s to be better value, personally)
 
3:01 AM
FWIW, Astaroth was designed with the possibility of SLI in mind, but as a less likely upgrade path rather than a primary use case.
 
well essentially phyx and gpu multi monitor vaguely are the same setup
just different uses of the secondary card
 
This can confuse games because they'd be using different hardware to render to different displays.
 
3:14 AM
As for power delivery... I have enough headroom in the power supply to support a second card, but barely; I chose 850W to ensure that there's plenty of headroom for peak loads and account for possible PSU degradation over time (which, frankly, would be very, very slow for a cream-of-the-crop 80 PLUS Titanium PSU) and, again, to allow for a possible SLI upgrade path, but I considered multi-GPU arrays an unlikely possibility considering just how powerful single cards can be these days.
(the 650W and 750W options were not particularly attractive, with the 650W model cutting it too close and the 750W model being a mere $20 less than the 850W model; a 1000W model was available but there was little advantage to going with such a large supply, with lower efficiency under typical loads and much higher cost)
As currently configured, absolute peak power is about 500W.
As tempting as the new RTX 20 cards are, I can't really justify replacing my existing, very expensive, and very rare GTX 1080 Ti FTW3 Elite card at this time.
7nm will bring another major advance and I'd rather wait until then.
 
Bob
3:34 AM
...RTX?
 
@Bob NVIDIA is promoting new ray-tracing hardware.
 
Bob
...
 
The fact that NVIDIA has dedicated hardware and middleware for ray-tracing capability is... unsettling, to say the least.
12 hours ago, by bwDraco
Looking forward to improved ray-tracing capability. AMD RTG is in deep trouble and I fear NVIDIA will force ray-tracing down developers' throats to shut down RTG...
yesterday, by bwDraco
NVIDIA has a big suite of proprietary technologies and even have a customized process at TSMC. AMD just does not have the money to match them.
This is making me genuinely concerned about the possibility of a discrete GPU monopoly.
NVIDIA has a history of inducing developers to implement technologies that disadvantage competitors. Considering how advanced a capability ray-tracing is, and the fact that it uses dedicated, fixed-function ASIC hardware in the Turing architecture, there is going to be no way AMD will be able to deliver comparable performance.
 
it is really cool tech ^^
 
As significant a technical advance this is, NVIDIA can use RTX as a cudgel to prevent AMD RTG from being able to compete effectively in all market segments above the entry level. The disadvantage that RTG would suffer from is potentially large enough to force them to exit the GPU market.
 
3:43 AM
but but.... most of the standards are openish
 
And with Intel's discrete graphics solution not due until 2020, this will result in a monopoly.
The only potential point of relief is that there's a ray-tracing API in DirectX, but I highly doubt AMD will be able to implement fixed-function ray-tracing hardware soon enough to be able to compete.
 
does Khronos have a ray tracing API yet?
 
*prays that Vega 7nm or otherwise AMD's next GPU architecture will have ray-tracing hardware*
 
implementing your own proprietary ray tracing fixed function hardware is probably a $500 million to $1B investment
where's AMD gonna get that?
 
This isn't just a matter of a new technology. It's something that is big enough to severely constrain AMD's ability to compete. This is why it's making me very fearful that a monopoly can result.
AMD's CPU division is doing just fine and is poised to overtake Intel in performance next year, but RTG is in serious trouble.
 
4:28 AM
You're constantly worrying about AMD not being able to compete ;p
@bwDraco apparently vulcan does
 
I suppose AMD could leverage some of its GPU compute prowess to implement ray-tracing...
 
@JourneymanGeek the way it sounds, he's worried about AMD not being able to compute :P
 
lol
 
AMD has a few... interesting tricks tho
like how they basically just sold an IP block to intel, or did multi package chips
 
4:31 AM
and they seem to win every console GPU except the Switch
 
@allquixotic poor IBM ;p
 
Switch is Tegra :(
 
switch is also a arm tablet arm
most of the current generation consoles are... oddly enough x86
 
In the end, it comes down to how many games will implement RTX. If it's a few games, it won't be a huge deal, but if a significant number of new titles use it (and NVIDIA has the ability to make that happen), RTG would be on deathbed.
AMD may be able to leverage a process advantage (GlobalFoundries 7LP vs TSMC 12FFN) to win in titles that don't use the new Turing features, but I'm not sure if that's going to be enough.
AMD needs to deliver 7nm Vega to consumers by yesterday.
(if you get what I mean)
 
here's a joke...
Basically Intel has had a processor that was literally perfect for raytracing... for years ;p
and have been working on it for ages
 
4:44 AM
Xeon Phi? Or even Larrabee?
This is why I've mentioned Intel's upcoming discrete GPU.
 
Larabee -> Knight's Corner -> Xeon Phi (Knight's Landing), etc... as I understand it rather inefficient but many integrated x86 cores is basically the architecture for raytracing IIRC
 
Which I suspect is why the RTX series was annouced first
@allquixotic or many arm CPUs... or even many RISC V Cpus
 
Ray-tracing is embarrassingly parallel. A second Radeon RX Vega card in a CrossFire configuration can perform this job with superb performance.
And AMD does have open-source middleware for this, called Radeon Rays.
The question is adoption and the ultimate capabilities of the underpinning hardware. NVIDIA's GPUs are without a doubt more efficient and more powerful than AMD's solutions. Whether any stopgap solution will allow them to remain viable for consumers is unclear.
 
also, new tech takes a while to catch on
 
4:50 AM
Yeah. AMD will have some time to develop a solution.
The question is, "can AMD deliver a comparable solution soon enough?"
So... I think that AMD RTG will hang in there but will hit rough waters in the coming months and into 2019. A new 7nm GPU could enable them to weather the storm and remain competitive.
 
ehhh
they have a few things in their favour
They sell IP Cores - and as far as video goes, they're the best in the market there
 
They have integrated CPU+GPU solutions that Intel (and indirectly, NVIDIA) is not able to compete with (at least at the price points in question). Stolas is living proof of this.
 
so unable to compete with that....
Intel bought an IP core off AMD ;)
 
And AMD makes highly-integrated semi-custom solutions that let them tap into unique embedded markets.
@JourneymanGeek Not quite a synthesizable IP core. The GPUs themselves, and a license to the Radeon driver stack.
 
@bwDraco I thought intel fabs those themselves
or is it a MCM?
 
4:59 AM
It's an MCM. The semi-custom 24-CU Vega GPU is manufactured by GlobalFoundries and Intel integrates it into the processor package.
Kaby Lake-G is an MCM consisting of an Intel H-type processor die, the aforementioned semi-custom Vega GPU (connected via PCIe), and a single, 4 GB stack of HBM2 (connected to the GPU via EMIB).
 
5:12 AM
I was thinking about this... how about a further cut-down version of Vega 10 with 48 CUs, tuned clock frequencies for higher efficiency while delivering mainstream (GTX 1060) performance, and a mixed 4 GB HBM2 + 4 GB GDDR5 configuration leveraging AMD's unique High-Bandwidth Cache Controller technology?
It would be cheaper to manufacture, taking advantage of their clever HBCC tech to reduce the amount of expensive HBM2 required without significantly hurting performance.
Heck, they could even omit the GDDR5 altogether on an entry-level solution.
At high clocks, 48 Vega CUs would perform like a GTX 1070, but AMD should focus on efficiency with this sort of part and clock it lower. Overclockers will have a field day with this if graphics card manufacturers would design cards around this GPU with beefy VRMs and cooling, but since this GPU would be aimed at the mainstream, AIB OEMs shouldn't use high-end power delivery with it.
(for comparison, Polaris 10 and Polaris 20 have 36 CUs)
 
5:38 AM
@bwDraco I feel like there's room on the chip to put way more CUs than there is either power or cooling to properly cool them if they all run sustained, but part of what I understand Max-Q does is it takes a bigger chip than could run full-bore in the system, and down-clocks it and throttles it appropriately so it can at least burst somewhat approaching its original capabilities
 
6:00 AM
Come to think of it... this hypothetical "Vega 48" GPU would be a perfect match for today's affordable 1080p144 FreeSync monitors. Here's an example:
Heck, this Acer monitor, which is on sale: amazon.com/Acer-XFA240-bmjdpr-Response-Technology/dp/B06ZYHZ6R6
The lower resolution (compared to, say, 1440p) would fit the smaller framebuffer with HBCC providing extra headroom for games that require a bit more than 4 GB of graphics memory. The GPU itself would deliver a quality 1080p gaming experience at silky-smooth frame rates.
 
6:16 AM
I have to do a ridiculous amount of studying today onwards for a month and a half. Ugh. Let's see how well I do.
 
6:28 AM
get stuck in it lad
 
 
4 hours later…
10:32 AM
@rahuldottech good luck, I'm in the same warpath. See you on the other side.
en.wikipedia.org/wiki/… <- Is there a similar page for India?
I got a listing of what's on TV timesofindia.indiatimes.com/tv/tvlistings
there ought to be a better way to get details than scraping
 
Bob
I'm covered in dog hair again
 
how do set-top boxes get synopsis and details of what's on?
@Bob makes sense if you have a molting doggo
 
11:28 AM
Magic
@Bob I don't shed much
 
>

India’s Official TV Guide App, covers 500+ Indian TV Channels!

“A must-have app for all TV junkies!” – The Times of India

"The best free app is Whats-On-India, because it’s fast and simple." – MINT



Quickly see what’s on TV, mark your favorite channels & shows, set reminders, and discover new shows.



Winner: Forbes Top Start-Ups to watch for, CNBC TV 18 Top 10 Start-Ups - 2010



The app features a neatly organized set of buttons to get you to the best movies, sports matches, TV shows and events happening on TV. This app puts at your fingertips a comprehensive program guide with th
Electronic program guides (EPGs) and interactive program guides (IPGs) are menu-based systems that provide users of television, radio and other media applications with continuously updated menus displaying broadcast programming (TV listings in the UK) or scheduling information for current and upcoming programming. Some guides also feature backward scrolling to promote their catch up content. They are commonly known as guides or TV guides. Non-interactive electronic program guides (sometimes known as "navigation software") are typically available for television and radio, and consist of a digitally...
 
12:21 PM
Ah, I think I can code up a set top box software with that info.
 
12:46 PM
@DavidPostill You should start charging for your Googling services
Hell, you could go on a freelancing website with "I can find anything on the internet for you!"
 
@rahuldottech i should use that.
 
 
3 hours later…
3:36 PM
@Bob you'll like this
> Securing the web literally made it less accessible to many, many people around the world.
> I think HTTPS is probably a net positive overall
> The drive to force every site on the web to HTTPS has pushed the web further away from the next billion users—not to mention a whole lot of the previous half-billion.
The article is just
Ugh
So
> a lab full of Windows XP machines limited to IE8, for example.
No, it's not. It's not, by any means
There's a ridiculous number of browsers which they can use
Is there a valid point to the article? Sortaaaa.
Also, I have no idea what service workers have to do with providing non-HTTPS access to websites?
> For users of up-to-date browsers, yes: service workers create a “good” man in the middle that sidesteps the HTTPS problem, so far as I understand.
YOUR WHOLE POINT WAS THAT THEY DON'T HAVE MODERN BROWSERS
 
there there.
 
And you absolutely set up a cache server for HTTPS websites ugh
 
Bob
@rahuldottech I think they're just missing that a cache is possible as long as you can explicitly trust the cache.
Fairly standard enterprise setup to have a filtering proxy that intercepts HTTPS.
All to need is an explicit trust on a custom root.
 
3:55 PM
Any Mexicans here?
 
no but we got some brazillian spanish speakers
 
There was one but we had to deport him, I think.
@jokerdino @ThatBrazilianGuy right :P
 
^-^
 
4:16 PM
One for @bwDraco featuring 2 noctua fans in a custom PC case (a nes).
 
@rahuldottech done! fiverr.com/s2/64e1a6bbd7
what do you guys need found?
 
@djsmiley2k I really want to build a Mini-ITX Raven Ridge desktop for my father.
Conventional? Unconventional? Ultra-compact case?
> ATX PSUs up to 160mm deep (shorter PSUs recommended for easier cable management)
Interesting...
(the Seasonic Focus Plus Gold line is 140mm long)
Well... since there is no need for a graphics card, I might be able to go even smaller than that.
(and considering that heavy gaming is unlikely to be necessary, discrete graphics may never be necessary)
 
4:40 PM
@Nick Illegal medicine
 
4:53 PM
@Nick Dude. You suck at googling. No offence. But you've shown a complete lack of looking-up-skills countless times.
ALSO GUYS LOOK I MADE SOMETHING TO MAKE PHP SESSIONS MORE MANAGABLE gist.github.com/rahuldottech/3d44b08f393bc5d4b2362fc03084b1d5
cc @bob
 
5:28 PM
0
Q: Audit reviews shouldn't treat comments as negative input

confettiSo I was reviewing first posts and came across a good question. Me thinking it has no answers yet, I left a comment asking for more information and suggesting a solution to the question. Afterwards, I wanted to upvote the question as it was good. However as soon as I left my comment I was greeted...

 
5:50 PM
@rahuldottech Please be a bit more careful with the language you're using. It reads like a personal attack and is borderline flaggable.
 
@rahuldottech aww, expected no lesser words of encouragement from you, bro.
 
6:03 PM
well, I've googled enough to find out you're no longer with your girlfriend. reddit.com/r/rarepuppers/comments/7xhimz/…
how's that for sucking at googling.
 
6:20 PM
@Nick everyone in this room knows that, I've stated it multiple times here
Like. That was almost a year ago. There's literally a public page I wrote on which it's stated. Everyone here knows about it.
And my Reddit profile is public, so... Big deal?
Also, that post says nothing about me being single or not
 
@Nick @rahuldottech Can you two please take your bickering somewhere else?
 
I'm sorry
 
Me too.
hugs?
 
Someone please explain to me why people ask programming questions on Quora. Like. Why. Just. Why.
 
6:27 PM
@rahuldottech They are question banned on Stack Exchange? ;)
 
@DavidPostill Never considered that possibility. Still. You can ask anonymously. Or make a new account. It's not hard.
 
It's not hard generally to find a solution to a problem either yet they still ask >_<
 
Creating new accounts to get around question/answer bans (suspensions) will get you into worse trouble if (when) you are caught ...
 
@DavidPostill it's really not that hard not to get caught
Also
> Worse trouble
Another ban which you can circumvent by creating yet another account?
I'm just saying. Don't break rules, kids
Speaking of which
A scene from my favorite show
 
 
1 hour later…
7:53 PM
...welp. replaced the embedded Twitter feed on my website, see twittercommunity.com/t/deprecating-widget-settings/102295
 
8:10 PM
You know what? I've disabled the whole thing because I'm not sure how to comply with the cookie rules.
My privacy policy also needs to be rewritten to accommodate this.
The code is still there but it is commented out.
It's always felt out of place and I don't really think it adds much value.
I'm also dealing with the need for HTTPS.
My website needs a major update...
 
 
1 hour later…
9:12 PM
The HP Support Assistant app that came with my ENVY x360 alerted me to a firmware update for my printer. It's a critical security update, so please update ASAP!
(CVE-2018-5924 and CVE-2018-5925; buffer overflow enabling remote code execution via crafted print data)
Applies to most HP printers of recent vintage.
 
Dog
9:33 PM
@allquixotic Ah nice. The one that has Android phone and notification integration?
Also, I've done a labelling SNAFU and mixed up my stimulant and sedatives bottles. I have no idea what's in which, and they both look the same, so I'm just going to keep taking this one until I either stop feeling tired or pass out
 
9:49 PM
@Dog Wat.
!!s/print data/fax/
 
@bwDraco (CVE-2018-5924 and CVE-2018-5925; buffer overflow enabling remote code execution via crafted fax) (source)
 
It's associated with the way color faxes are processed.
 
10:20 PM
@Dog idk, just in general
someone posted it to chat a little while ago
@Dog jesus, dude, that's super dangerous
 
 
1 hour later…
Dog
11:21 PM
@allquixotic Like most things in my life right now
Actually tbh, while it's not smart, it's not exactly dangerous. I've had more than what's left in this bottle in one go, and been fine
 
@Dog ok. So. Folks have expressed discomfort with both the drugs and the talk of it. I really am going to have to ask you to stop
3
 
HAMMERTIME!
 
.... Damn it cavil...
 
Dog
But smoking, drinking coffee, and getting drunk are all taking drugs!
 
Not the same in this context and you know it
 
Dog
11:35 PM
You just don't like my drugs cause we have an NHS that provides them for free and you don't.
 
No. I don't like the story that you're taking it in non standard ways and doing various silly things under the influence. And people are uncomfortable about that
So. This stops now
 
Dog
I don't like you but I tolerate you anyways
Would you rather hear about my lube stories instead? Cause there's not much else going on in my life right now.
 
This... is degenerating into trolling.
 
No
That makes me uncomfortable. Personally
 
11:44 PM
Meh. No real need
 
Dog
You know what makes me uncomfortable? Taylor swift.
 
@Dog so make more of it than lubes and drugs
 
HAMMERTIME!
 
@Dog why?
 
Dog
She's just too... sassy
 
11:47 PM
Erm. Excellent. You can have a reasoned conversation with the next person who causes a ts gif or image to be posted on channel to why sassy women make you uncomfortable and they should stop
 
HAMMERTIME!
 
Dog
!!taytaytay
 
Dog
Hey it's not sassy women in general, that's stereotyping.
 
.... See, if it makes you uncomfortable you wouldn't have done that.
 
Dog
11:50 PM
Also, me? Having a reasoned conversation about anything other than phone masts and RAN configurations? Dream on.
 
This has degenerated into personal attacks.
 
Those things are fine, and well appreciated
 
And no, we do not appreciate your trying to cover your tracks by editing the message to remove the personal attack.
 
@bwDraco not really, and let me deal with this please?
 
Dog
Oh and I've managed to go from making $300 profit to losing $1700 on cryptocurrency. The very opposite of reasoned
 
11:56 PM
Well crypocurrency is hardly reasonable. But we're talking about your interactions with others, not your hobby of virtual tulip trading
And if haven't heard any complaints about that, except when you talk about doing it under the influence
 

« first day (2935 days earlier)      last day (2085 days later) »