Author Topic: Soap router - HW specs too good to be true?  (Read 97550 times)

0 Members and 1 Guest are viewing this topic.

Offline HackedFridgeMagnet

  • Super Contributor
  • ***
  • Posts: 2028
  • Country: au
Re: Soap router - HW specs too good to be true?
« Reply #100 on: July 10, 2014, 04:00:08 am »
No, I think US law applies world wide.  8)
 

Offline Legit-Design

  • Frequent Contributor
  • **
  • Posts: 562
Re: Soap router - HW specs too good to be true?
« Reply #101 on: July 10, 2014, 04:04:59 am »
Does he realise that the owner of the forum is actually not living in the USA? As well he will have to prove this to each jurisdiction world wide.
I think this forum and other sites are hosted by hostgator in texas USA? I don't think that is secret knowledge? And USA is the leader of the free world and leader of the internets. But anyways since something like the piratebay cannot be closed down from the internet, good luck trying to close this site down. Which has mostly healthy critique and some people who actually know their shit. Why attack this site if it's not actually hurting them in a real way. Truth hurts...
 

Offline dext0rb

  • Contributor
  • Posts: 49
  • Country: us
  • grind it out
Re: Soap router - HW specs too good to be true?
« Reply #102 on: July 10, 2014, 04:38:16 am »
Um, slander is spoken defamation. I think he's looking for 'libel'. Anyways, what a tool. Wonder if they sent a similar threat to hackaday?
EE and Tinkerer.
 

Offline Bored@Work

  • Super Contributor
  • ***
  • Posts: 3932
  • Country: 00
Re: Soap router - HW specs too good to be true?
« Reply #103 on: July 10, 2014, 05:05:23 am »
For the public record:
The Soap CTO put in an abuse report against this forum to my ISP (well, cloudflare actually) claiming:
Quote
I will be filing defamation and slander suits to each contributor and person who has taken part in this.  They continue to endorse it and to attack us personally, our company and our project.

That would require:

A) They actually have a CTO?

I doubt they have one, given the lack of technology knowledge. And I couldn't find anyone billing himself as the CTO on the two crowdfunding campaign pages.


B) There is actually a corporate entity, presumably called Soap Inc.?

Trying to look for Soap Inc. and their corporate officers I came up with nothing. (Ok, I found a Soap Inc producing real soap, as stuff for washing). Does the router Soap Inc. company exist at all?


C) The CTO is authorized to make legal threads in the name of the company?

Typically companies don't allow their officers, especially not the T, to run around and make legal threads. They send the corporate lawyer, or the CEO makes statements, carefully crafted by the corporate lawyer.


Taking A), B), and C) into account I think that "Soap CTO" is full of shit.
« Last Edit: July 10, 2014, 04:15:31 pm by Bored@Work »
I delete PMs unread. If you have something to say, say it in public.
For all else: Profile->[Modify Profile]Buddies/Ignore List->Edit Ignore List
 

Offline Smokey

  • Super Contributor
  • ***
  • Posts: 2536
  • Country: us
  • Not An Expert
Re: Soap router - HW specs too good to be true?
« Reply #104 on: July 10, 2014, 05:18:32 am »
.....Wonder if they sent a similar threat to hackaday?
That's a good point.  Might be why the tone turned so positive all of a sudden. 
Guess we'll have to wait and see what Brian from hackaday comes up with since it's coming up on "a week or two".
http://hackaday.com/2014/06/26/the-ifind-kickstarter-campaign-was-just-suspended/comment-page-1/#comment-1598914

Name and shame the alleged "CTO" maybe?
 

Offline rob77

  • Super Contributor
  • ***
  • Posts: 2085
  • Country: sk
Re: Soap router - HW specs too good to be true?
« Reply #105 on: July 10, 2014, 06:55:49 am »
don't you guys have some kind of official company/business directory in the US ? someone should look up them there.
if there is no official company registered under SOAP Inc. , then it's an illegal business i think ;) (at least it would be illegal here)

over here we have a official list for all businesses - corporations, limited liability, personal business licenses - everything is online, so you can look-up anything online. and of course all persons responsible for the company/business are listed there as well - so once you run your business here, you can't hide from public ;)
 

Online EEVblog

  • Administrator
  • *****
  • Posts: 37661
  • Country: au
    • EEVblog
Re: Soap router - HW specs too good to be true?
« Reply #106 on: July 10, 2014, 07:31:13 am »
BTW, if Soap are reading, then you are more than welcome to register on the forum and explain your project and respond to any criticism.
Reporting the forum to the ISP is not cool (and pointless), but engaging with the technical community is a very cool thing to do. Countless examples of companies doing that on here and being much better off for it.
 

Offline HackedFridgeMagnet

  • Super Contributor
  • ***
  • Posts: 2028
  • Country: au
Re: Soap router - HW specs too good to be true?
« Reply #107 on: July 10, 2014, 11:05:41 am »
But lets face it, they are scammers. Their best bet is to ignore us and try to get as much money as they can.
 
 

Offline madires

  • Super Contributor
  • ***
  • Posts: 7695
  • Country: de
  • A qualified hobbyist ;)
Re: Soap router - HW specs too good to be true?
« Reply #108 on: July 10, 2014, 12:55:02 pm »
Good luck serving those summonses, though the sheriff of the local court who has to serve them will love the worldwide travel and accommodation that he will have to pay to serve.

Does he realise that the owner of the forum is actually not living in the USA? As well he will have to prove this to each jurisdiction world wide.

I think they are just trying to silence any critics by intimidation. Our critic is technically well grounded and can be proven in front of any judge. Anyway, most stuff written here is protected by free speach according to US law and the laws of a lot of other countries, as long as it's reasonable. If they really want to file suits they'll face a huge risk in loosing a lot of money and any lawyer would tell them that. The backers won't be happy if the Soap team would burn the money for pointless law suits, might be embezzlement. IIRC the US DOJ has filed the first suit against a crowd funded project that didn't deliver.
 

Offline dext0rb

  • Contributor
  • Posts: 49
  • Country: us
  • grind it out
Re: Soap router - HW specs too good to be true?
« Reply #109 on: July 10, 2014, 07:01:56 pm »
Quote
With Soap mesh, you can create a 3 GB/s wireless link.

I'm going to sue them for making false and/or misleading claims. Show me some performance tests & data on your 3 Gigabyte per second wireless link.  Units matter!
EE and Tinkerer.
 

Offline echen1024

  • Super Contributor
  • ***
  • Posts: 1660
  • Country: us
  • 15 yo Future EE
Re: Soap router - HW specs too good to be true?
« Reply #110 on: July 10, 2014, 07:25:48 pm »
Just noticed this on their website. And they wonder why people don't take them seriously.

I'm not saying we should kill all stupid people. I'm just saying that we should remove all product safety labels and let natural selection do its work.

https://www.youtube.com/user/echen1024
 

Offline Hardcorefs

  • Regular Contributor
  • *
  • Posts: 81
Re: Soap router - HW specs too good to be true?
« Reply #111 on: July 11, 2014, 02:01:14 pm »
For the public record:
The Soap CTO put in an abuse report against this forum to my ISP (well, cloudflare actually) claiming:
Quote
I will be filing defamation and slander suits to each contributor and person who has taken part in this.  They continue to endorse it and to attack us personally, our company and our project.
:-// We are criticizing your project. If you can't take criticism, don't get on the internet.

I think the issue is that looking at some of the posters on EEVblog, they are using words such as 'scam' and 'faked',  It should be possible for people to take a look at the design and make  valid points without resorting to such terms.

For example we could be looking at :
The placement of 'osc5/osc6'  6 seems most likely to clash with the metal case on the RJ45 pins after assembly, then there is the clash with the SMT on the right of osc6.( not to mention on most of the giga routing chips it specifically states to keep ALL oscillator circuits away from the data tracks

The 'differentials' for the  giga internet, at those speeds every 45 Deg. angle in a PCB track is potentially a dangerous curve.

As regards  peoples comment to the software and it being difficult to write,  'Android' has a linux foundation , there are already a multitude of 'firewalls', routers & switches already in the public domain. ( personally I would NOT have used Android' any where near this, all the GUI can be implemented in a far cleaner way)

However a linux SBC does not a router/switch/firewall make…., for such a design to be truly effective,  packet switching MUST be implemented with a switching fabric, slapping 'giga' rate packets between ports is not something you do by routing the crap via an ARM CPU.

In fact I cannot find any mention of the word 'switch' any place in the kick-starter blurb… which would get me thinking that perhaps this is nothing more than a packer 'router', where each packet is forwarded to each and every connection in a dumb manner.
(generally Switches intelligently switch packets based on the headers and ONLY forward packets to where they are supposed to go, routers just shove every packet over every endpoint with NO regard for the header…)

Which has the potential to be a massive security issue, since that would allow any connection to see the packets of any other connection, a single PC loaded with 'wireshark' on a router based network is the kiss of death, and another reason NOT to build a firewall that way…

RF
« Last Edit: July 11, 2014, 02:06:30 pm by Hardcorefs »
 

Offline madires

  • Super Contributor
  • ***
  • Posts: 7695
  • Country: de
  • A qualified hobbyist ;)
Re: Soap router - HW specs too good to be true?
« Reply #112 on: July 11, 2014, 03:26:35 pm »
In fact I cannot find any mention of the word 'switch' any place in the kick-starter blurb… which would get me thinking that perhaps this is nothing more than a packer 'router', where each packet is forwarded to each and every connection in a dumb manner.
(generally Switches intelligently switch packets based on the headers and ONLY forward packets to where they are supposed to go, routers just shove every packet over every endpoint with NO regard for the header…)

Some network 101 for you ;) A classic switch is a layer 2 device, e.g. switching Ethernet frames based on the MAC address of the destination. The common SOHO router with 4 LAN ports got an Ethernet switch for the LAN ports and 5th port is connected to the CPU. Switching IP packets (also called layer 3 switching) is something completely different. The normal process of moving packets around is called forwarding. The router reads the destination IP address, looks up that address in the routing table and then forwards the packet to the next hop. Layer 3 switching is a method to offload the CPU from checking the routing table for each packet. There are several methods available for layer 3 switching, some are completely software based, and others use a mix of software and ASICs. There are also some solutions using sub-processors which perform forwarding and layer 3 switching. Anyway, you won't find any SOHO router supporting layer 3 switching, it's to expensive for that kind of hardware and market.
 

Offline dext0rb

  • Contributor
  • Posts: 49
  • Country: us
  • grind it out
Re: Soap router - HW specs too good to be true?
« Reply #113 on: July 11, 2014, 04:13:57 pm »
For the public record:
The Soap CTO put in an abuse report against this forum to my ISP (well, cloudflare actually) claiming:
Quote
I will be filing defamation and slander suits to each contributor and person who has taken part in this.  They continue to endorse it and to attack us personally, our company and our project.
:-// We are criticizing your project. If you can't take criticism, don't get on the internet.

I think the issue is that looking at some of the posters on EEVblog, they are using words such as 'scam' and 'faked',  It should be possible for people to take a look at the design and make  valid points without resorting to such terms.

I see your point, but it is very hard not to feel the Soap people were being a bit disingenuous when they tried to pass off bunnie's Novena PCB as their own prototype.
If you look at one of the initial "prototype" videos, they have a hard drive positioned to look like it is connected to the eval board - but the board itself has no SATA ports...

The foundation of their "relationship" (if you want to call it that) with backers and/or The Internet is mired in what appears to be dubious behavior.
I believe it will be hard to evade the "scam" or "fraud" tag now, based on said "relationship". Instead of responding to technical criticism with technical info, they come out swinging with legal threats, which seems to me, would be the response of someone who is not qualified to answer those technical criticisms.
EE and Tinkerer.
 

Offline Rasz

  • Super Contributor
  • ***
  • Posts: 2616
  • Country: 00
    • My random blog.
Re: Soap router - HW specs too good to be true?
« Reply #114 on: July 11, 2014, 07:01:35 pm »
I think the issue is that looking at some of the posters on EEVblog, they are using words such as 'scam' and 'faked'

Because it is a scam. Few guys with "good idea" and ZERO knowledge about implementation and no hope of delivering duck make Scampaign using FAKE art assets. Lying about Bunnies laptop pcb being their own work, using clever angle/crop shots normally employed by fat chicks on myspace to pretend their prototype actually exists. Then receive $xxxK - this automagically:
-makes them think they have something real
-that this money will somehow design product for them, just pay subcontractors, right?
-lures clueless VC money

VC will drop few mill on their laps, because VC is just as retarded as those retards and 'money will somehow design product for them'. At this point they are in 'just push something out the door at a loss, we will make it up in volume later' phase, except we have so much money we can start adding features ....
for example
http://arstechnica.com/gadgets/2014/07/how-one-kickstarter-project-squandered-3-5-million/



For example we could be looking at :
The placement of 'osc5/osc6'  6 seems most likely to clash with the metal case on the RJ45 pins after assembly, then there is the clash with the SMT on the right of osc6.( not to mention on most of the giga routing chips it specifically states to keep ALL oscillator circuits away from the data tracks

meh, thats just pcb semantics, will work on third board spin ($xxK spend at that point) if particular subcontractor that did that carrier board for them has a clue. Electronics are not rocket science for this project. Electronics at _particular price point_ accompanied by appropriate software stack is the key.
I can deliver Soap router today .. by shipping $1500 laptop to every backer.

The 'differentials' for the  giga internet, at those speeds every 45 Deg. angle in a PCB track is potentially a dangerous curve.

dude cmon, not this 'hi speed electrons will fall off at the turn' bullshit

As regards  peoples comment to the software and it being difficult to write,  'Android' has a linux foundation , there are already a multitude of 'firewalls', routers & switches already in the public domain. ( personally I would NOT have used Android' any where near this, all the GUI can be implemented in a far cleaner way)

Android ripped stock Linux network stack, there is no iptables, no netfilter, no nftables.

for such a design to be truly effective,  packet switching MUST be implemented with a switching fabric, slapping 'giga' rate packets between ports is not something you do by routing the crap via an ARM CPU.

sounds kewl in your head, sounds retarded when you say 100MB/s stream cant be parsed by 1GHz of processing power (times 4 in quad core). Thats ~10K cpu cycles per packet. Even shittiest cheapest desktop board with Pentium cpu can route at line speed. If you need performance the trick is routing in userspace, bypassing linux stack overhead (this is how hi frequency trading people handle all their network traffic on linux boxes).

In fact I cannot find any mention of the word 'switch' any place in the kick-starter blurb… which would get me thinking that perhaps this is nothing more than a packer 'router', where each packet is forwarded to each and every connection in a dumb manner.
(generally Switches intelligently switch packets based on the headers and ONLY forward packets to where they are supposed to go, routers just shove every packet over every endpoint with NO regard for the header…)

someone already explained that one :)

Anyway, you won't find any SOHO router supporting layer 3 switching, it's to expensive for that kind of hardware and market.

https://www.turris.cz/en/
L1-3 full hw acceleration.




Waiting for hackaday update, at this point its pretty obvious legal (well, not so legal when made by a fictitious corporate persona) threats were the reason of attitude change. VC( if they have any) is more than happy to pay lawyer fees to kill all the bad PR while its fishing for a bigger sucker to buy this 'disruptive innovative in teh CLOUD next NEST' thing off their hands.
Who logs in to gdm? Not I, said the duck.
My fireplace is on fire, but in all the wrong places.
 

Offline dext0rb

  • Contributor
  • Posts: 49
  • Country: us
  • grind it out
Re: Soap router - HW specs too good to be true?
« Reply #115 on: July 11, 2014, 07:08:00 pm »

The 'differentials' for the  giga internet, at those speeds every 45 Deg. angle in a PCB track is potentially a dangerous curve.

dude cmon, not this 'hi speed electrons will fall off at the turn' bullshit

+1
EE and Tinkerer.
 

Offline rob77

  • Super Contributor
  • ***
  • Posts: 2085
  • Country: sk
Re: Soap router - HW specs too good to be true?
« Reply #116 on: July 11, 2014, 08:56:38 pm »
(generally Switches intelligently switch packets based on the headers and ONLY forward packets to where they are supposed to go, routers just shove every packet over every endpoint with NO regard for the header…)

 :-DD sorry man.... but i couldn't help myself... i would give you a networking nobel prize for this one ;)
no offense - but you should really learn some stuff around networks. start with wikipedia articles and then read some textbooks (e.g. cisco CCNA courses are good to start learning networks).

but in general:
HUB - layer2 -  sends everything everywhere
SWITCH - layer2 - switching ethernet frames based on MAC table (MAC to port bindings) - if the MAC is not in the table - sends it everywhere
ROUTER - layer3 - forwards packets to particular destinations based on routing table
LAYER3 SWITCH - special case it's a L2 switch and a L3 router combined to one device - the routing is done in hardware (super fast for plain routing), but it has not enough CPU power for advanced stuff like GRE tunnels..etc..
for example the good old Catalyst3550 with L3 firmware (layer3 switch) was able to route a shitload of traffic while handling a routing table with 2k+ prefixes (BGP) , but the CPU become dead slow during 10Mbps traffic via a GRE tunnel (terminated on that catalyst).

btw.. regarding SOHO routers - in many cases of the soho routers (especially the cheap ones with lot's of features) have a 6 port switch with primitive 802.1q capabilities - 4 ports available for LAN connection , 1 port for WAN and the last port connected to the CPU is a VLAN trunk carrying both LAN and WAN traffic (each with a different tag). Wifi on those is usually connected to the CPU
 

Offline BartManInNZ

  • Contributor
  • Posts: 46
  • A Geek In New Zealand
    • YouTube Channel
Re: Soap router - HW specs too good to be true?
« Reply #117 on: July 12, 2014, 07:28:05 am »
Their latest update about the wireless takes the cake!
"If you want to build a ship, don't herd people together to collect wood
and don't assign them tasks and work,
but rather teach them to long for the endless immensity of the sea." - Antoine de Saint Exupery
 

Offline Hardcorefs

  • Regular Contributor
  • *
  • Posts: 81
Re: Soap router - HW specs too good to be true?
« Reply #118 on: July 12, 2014, 08:10:10 am »

The 'differentials' for the  giga internet, at those speeds every 45 Deg. angle in a PCB track is potentially a dangerous curve.

dude cmon, not this 'hi speed electrons will fall off at the turn' bullshit

+1

It is nothing to do with the electrons, rather the track is smother for the  '1' & '0',  It is not difficult to understand.
Get a  flexible pipe and drop some marbles and spaghetti down it, now bend the tube at 45 degrees and see what happens.


 

Offline fubar.gr

  • Supporter
  • ****
  • Posts: 365
  • Country: gr
    • Fubar.gr
Re: Soap router - HW specs too good to be true?
« Reply #119 on: July 12, 2014, 10:17:22 am »
It is nothing to do with the electrons, rather the track is smother for the  '1' & '0',  It is not difficult to understand.
Get a  flexible pipe and drop some marbles and spaghetti down it, now bend the tube at 45 degrees and see what happens.

But electrons are neither marbles nor spaghetti. At high frequencies electrons vibrate, but stay put. What moves around is the EM field generated around the transmission line.

Offline madires

  • Super Contributor
  • ***
  • Posts: 7695
  • Country: de
  • A qualified hobbyist ;)
Re: Soap router - HW specs too good to be true?
« Reply #120 on: July 12, 2014, 11:26:46 am »
The wifi update is full of verifiable lies. It won't double the speed to your device or offer SATA like data transfer rates, because no device has two wifi radios that can be bonded together to double the bandwidth. Also, $100 for a shitty Marvell 802.11acbgn card? You can buy them for about $15 each. That's Apple levels of gouging.

Link aggregation on layer 2 requires links with the same speed. Since WLAN connections change the link speed all the time they can't be used for link aggregation. The obvious solution would be to use IP load sharing, but that requires a sophisticated algorithm for links with different and changing speeds. Another issue with available IP load sharing solutions is, that they can't fully utilize the links. The utilization depends on the mix of IP connections. If someone develops an algorithm solving all those issues he would get a nice job offer from Cisco or Juniper ;) For me, it's just more marketing hogwash.
 

Offline dext0rb

  • Contributor
  • Posts: 49
  • Country: us
  • grind it out
Re: Soap router - HW specs too good to be true?
« Reply #121 on: July 12, 2014, 06:41:38 pm »

The 'differentials' for the  giga internet, at those speeds every 45 Deg. angle in a PCB track is potentially a dangerous curve.

dude cmon, not this 'hi speed electrons will fall off at the turn' bullshit

+1

It is nothing to do with the electrons, rather the track is smother for the  '1' & '0',  It is not difficult to understand.
Get a  flexible pipe and drop some marbles and spaghetti down it, now bend the tube at 45 degrees and see what happens.

I believe Rasz was being facetious - of course electrons don't "fall off", but when you say that a 45 degree bend is a "dangerous" route - I don't buy it.

Even a "worst-case" 90 degree bend has almost no impact on a signal with a rise-time of 17ps, according to this paper. It also mentions your "not difficult to understand" electrons are marbles (or spaghetti??? lol  :) ) analogy. I do not believe this analogy is accurate.

This Gigabit transceiver shows a min rise-time of 150ps on the differential output.

At higher and higher frequencies, yes, I think the geometry of the trace plays more and more into any impedance discontinuities, but to start worrying about 45-degree bends on GigE...I think that is a bit rich. 
« Last Edit: July 12, 2014, 06:46:59 pm by dext0rb »
EE and Tinkerer.
 

Offline cyr

  • Frequent Contributor
  • **
  • Posts: 252
  • Country: se
Re: Soap router - HW specs too good to be true?
« Reply #122 on: July 12, 2014, 07:39:37 pm »

This Gigabit transceiver shows a min rise-time of 150ps on the differential output.

That's for 802.3z (1Gbps over a single signal pair with two voltage levels). 802.3ab (regular copper GigE) is much much slower, it uses 5-level encoding and all four cable pairs in parallel to get the data rate and is really robust. It's designed to go over 100m of cheapo cat5 cable, a couple of cm on the PCB at an odd angle isn't going to make it fail...
 

Offline dext0rb

  • Contributor
  • Posts: 49
  • Country: us
  • grind it out
Re: Soap router - HW specs too good to be true?
« Reply #123 on: July 12, 2014, 07:59:28 pm »

This Gigabit transceiver shows a min rise-time of 150ps on the differential output.

That's for 802.3z (1Gbps over a single signal pair with two voltage levels). 802.3ab (regular copper GigE) is much much slower, it uses 5-level encoding and all four cable pairs in parallel to get the data rate and is really robust. It's designed to go over 100m of cheapo cat5 cable, a couple of cm on the PCB at an odd angle isn't going to make it fail...

Word, thanks for pointing that difference out.
EE and Tinkerer.
 

Offline EricVanWyk

  • Contributor
  • Posts: 10
Re: Soap router - HW specs too good to be true?
« Reply #124 on: July 12, 2014, 08:10:38 pm »

I believe Rasz was being facetious - of course electrons don't "fall off", but when you say that a 45 degree bend is a "dangerous" route - I don't buy it.

Even a "worst-case" 90 degree bend has almost no impact on a signal with a rise-time of 17ps, according to this paper. It also mentions your "not difficult to understand" electrons are marbles (or spaghetti??? lol  :) ) analogy. I do not believe this analogy is accurate.

This Gigabit transceiver shows a min rise-time of 150ps on the differential output.

At higher and higher frequencies, yes, I think the geometry of the trace plays more and more into any impedance discontinuities, but to start worrying about 45-degree bends on GigE...I think that is a bit rich. 


Thanks for the link!

I had someone (who claimed to be an engineer) tell me that Ethernet was too unreliable, because:
  • If the cable is longer than 100 meters, you will blow the amplifiers.
  • If the cable is shorter than 1 meter, you will blow the amplifiers.
  • If the cable has a bend with a radius of less than an inch, the packets can't make it through.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf