We are nobody’s authorized dealer. That is deliberate.
No manufacturer pays us, certifies us, or sets our margins. Below is what we install, grouped by system, including the parts of each product line you would rather hear about now than discover in year three.
No dealer status, no partner status, no manufacturer certifications
Gotham Access Control NY is not an authorized dealer, certified partner, master dealer, elite installer or anything similar for any manufacturer named on this page or anywhere else on this site. We hold no manufacturer certifications and we display no manufacturer logos.
That statement is legally necessary (claiming a dealer relationship that does not exist is a straightforward misrepresentation), and it is also the most useful thing on the page. Equipment gets specified here on merit and on what your building can physically carry, not on which volume tier we are trying to hit this quarter.
What being unaffiliated costs you, honestly
Dealer agreements are not sinister, but they are commercial. They come with volume commitments, margin tiers, rebate thresholds and training requirements, and a contractor deep into one manufacturer’s program has a real financial reason to arrive at the same recommendation regardless of what is in your riser. That is why so many buildings across this city end up with an identical panel from an identical brand in wildly different circumstances.
The flip side is genuine and you should know it. Being unaffiliated means no factory technical-support escalation path, no dealer-only pricing tier, and occasionally longer lead times because we buy through distribution like everyone else. Where a manufacturer program would actually benefit you (extended warranty terms that only flow through an authorized channel, for instance), we will tell you, and if that is the deciding factor we will tell you to use an authorized dealer for that specific product.
What you get in exchange is a recommendation with no hidden term in it, and the ability to ask “why not the other one?” and get a real answer.
Intercom and entry
This is the category where the building decides the product, not the other way round. The right answer in a 1926 walk-up with a tired riser is frequently not the flagship.
Aiphone
The default for multi-tenant work in New York, and usually for good reason. The GT series is an analog bus platform explicitly designed for retrofit: it scales to hundreds of entrance stations and thousands of tenant stations and it runs on building wiring rather than demanding a home run to every apartment. IX is the IP commercial platform, IXG the IP multi-tenant residential one with a cloud tier and mobile app users in the thousands, and the JO, JP and JV series cover townhouses, small commercial and single-family. Strengths: an enormous installed base, real documentation, parts you can actually get, and no compulsory subscription on the analog lines. Weaknesses: GT modular entrance panels get expensive quickly once you add a camera module, a digital directory and a reader; the app experience trails the cloud-native competitors; and there is a specific field trap: Aiphone specifies polyethylene-insulated cable for the bus, and the PVC-jacketed cable a tech grabs off the truck has higher capacitance and will degrade the bus over distance.
Comelit
The two-wire specialist. Simplebus and Simplebus 2 multiplex audio, video, power and control over a single polarity-insensitive pair, with line distributors branching the riser; Simplebus 2 runs roughly 200 meters between outdoor and indoor units and the Top tier scales dramatically further. Strengths: often the cleanest way to get video over two existing conductors, and you can stage a migration floor by floor rather than in one cutover. Weaknesses: US documentation and support are thinner than Aiphone’s, and fewer installers in this market know it well, which matters in five years when you need somebody other than us to service it.
Akuvox
Broad hardware catalog, aggressive feature-per-dollar, and notably a two-wire IP line positioned specifically to replace legacy two-wire systems, which is a genuinely useful thing to exist in a city full of them. Strengths: capability at a price the incumbents do not match. Weaknesses: firmware quality has varied between releases, and the cloud tier is proprietary, so a meaningful part of what you are buying is a vendor’s app rather than the hardware on the wall.
2N
Czech, owned by Axis, and the choice when diagnosability matters. The IP Verso platform is a true SIP endpoint with modular readers, Bluetooth and fingerprint options, plus the part nobody markets but every technician values: syslog and packet capture built into the device’s own web interface. When a 2N unit misbehaves you can find out why instead of guessing. Strengths: open SIP means you are not married to one manufacturer’s app; build quality is excellent. Weaknesses: expensive, modules add up fast, and SIP is only an advantage if you have a phone system or platform to register it to. In a 12-unit walk-up it is over-specified.
ButterflyMX
Cloud, mobile-first, and architecturally different from everything above: a lobby panel, a resident app, and no in-unit hardware or apartment riser at all. In a building whose riser is genuinely dead, that is not a feature, it is the entire commercial argument. It deletes the most expensive line in the estimate. Strengths: tenant turnover becomes a web portal edit instead of a truck roll and a re-lettered directory; time-limited visitor codes and delivery access are handled properly. Weaknesses, and they are real: a per-unit subscription that never ends and can be repriced by the vendor; a closed platform with no meaningful exit; and one panel as a single point of failure for every apartment’s entry, which makes surge protection and a UPS mandatory rather than optional. One installation detail that catches people: the panel relays must not switch a strike coil directly. An isolation or buffer relay goes between them, or inductive kickback destroys the contacts.
DoorBird
Well-built German hardware, IP65, wide-angle lens, PoE or low-voltage DC, and an open API that integrates cleanly with Control4, Crestron and the usual home automation platforms. Strengths: excellent for a townhouse, a brownstone or a small office. Weaknesses: no built-in SIP, and a hard ceiling of eight mobile devices per door station. It is a single-family and very-small-building product regardless of what the multi-tenant keypad model suggests, and specifying it for a 30-unit building is a mistake you will discover at unit nine.
DoorKing
The workhorse of gate and telephone entry. The 1835 stores thousands of phone numbers and device codes and expands to dozens of access points; the 1837 can be programmed remotely. Strengths: rugged, repairable, and well understood by anyone who has worked on gates. The weakness is not the hardware at all. It is that the entire category was designed around a copper phone line that carriers are retiring. The signature symptom is a resident answering the call, pressing 9, and nothing happening, because the tone has been transcoded away somewhere in a VoIP path. The fixes are a cellular communicator that presents a simulated dial tone to the panel, or replacement with a native cellular or IP unit.
Alpha Communications and TekTone
The legacy amplifiers behind a large share of the audio intercoms in this city. Strengths: repairable, inexpensive, thoroughly understood, and frequently the right answer is to fix one rather than sell a building a new system. Weaknesses: audio only, no logging, no remote administration, and a shared-bus architecture where one water-damaged apartment station can load the bus and kill audio building-wide, a fault that has to be found by halving the riser at the distribution point until the bad leg appears.
Also specified
Latch (now DOOR) for cloud entry with Ethernet, WiFi or cellular connectivity, noting that the core intercom has no built-in camera. Siedle for premium modular architectural work where the entrance is part of the design and cloud dependency is unwelcome. Mircom where the building wants entry tied into a wider life-safety estate, with the fire alarm side handled by the appropriate contractor. Linear and Enterphone telephone-entry equipment, mostly encountered on service rather than specified new.
What your existing riser lets you buy
The survey answers this. Until it does, everything else is a catalog conversation.
| What is in the building now | The realistic path | The trade you are making |
|---|---|---|
| Legacy audio riser, most conductors sound | Modern audio or two-wire video over the existing cable: Aiphone GT, Comelit Simplebus 2, or Akuvox two-wire IP | Cheapest and least disruptive by a wide margin. You inherit the existing cable’s condition, and any future fault lives in a shaft you chose not to replace |
| Legacy riser, too few usable pairs or insulation breaking down | New riser cable, or a cloud panel with no in-unit hardware at all | Riser replacement is the expensive, durable answer. The cloud panel avoids it entirely and converts capital cost into a subscription that never stops |
| Legacy coax video riser | Reuse the coax with Ethernet-over-coax converters, or migrate to IP on new cable | Coax reuse preserves a large share of the original installation cost. It also caps how far the system can go later |
| Working Cat5e or Cat6 to every unit | Full IP: Aiphone IXG, 2N, Comelit IP | The most capability and the best diagnostics available. Only realistic where the structured cabling genuinely already exists |
| Telephone entry panel on a copper line | Cellular communicator presenting a simulated dial tone, or a native cellular or IP panel | The communicator is cheap and buys years. Replacement removes the dependency and adds a directory you can edit without a site visit |
Every row here assumes the entrance door, closer and strike are serviceable. On pre-war buildings that assumption fails often enough that it is checked at the survey, not at commissioning.
Access control, locking hardware and power
HID and Mercury controllers
HID Mercury boards are the open-architecture backbone of the category and sit inside a surprising number of systems sold under other names. A two-door intelligent controller handles readers, supervised inputs and relay outputs with downstream expansion; the higher-tier board adds native elevator destination-dispatch support. Strengths: because the same hardware runs under multiple head-end platforms, you can change software vendors later without replacing panels, the strongest anti-lock-in position available in access control, and worth real money in year seven. HID’s Signo readers cover Apple Wallet, Google Wallet, MIFARE, DESFire and Seos credentials on one device, which makes phased credential migration practical. Weaknesses: it needs a head end and a competent integrator behind it, and for four doors in a small office it is more system than the problem deserves.
Brivo
Cloud, and deliberately vendor-neutral about what it connects to: third-party video, directory services and HR systems all integrate rather than being locked out. Notable for multifamily buyers: pricing runs on a per-unit axis rather than per-door, which changes the arithmetic completely on a 200-unit building. Weaknesses: the per-door tiers are steepest at the small end, so a two-door site pays the highest rate per door of anyone; and the subscription is permanent operating expenditure that will be repriced over the life of the system.
Verkada
Fast to deploy, genuinely good administration, and a hybrid architecture where controllers keep enforcing the last known access list during an internet outage, a question every buyer should ask and few do. Supports proximity, encrypted smart cards, mobile NFC and Bluetooth, and plate recognition as a credential. Weaknesses: proprietary end to end. The controllers currently support the encrypted reader protocol only with Verkada’s own readers, so “open standard” does not mean what it sounds like here, and annual license renewal is the business model. Model the five-year number before signing, not the hardware price.
Avigilon Alta, LenelS2, Genetec and Kantech
Alta (formerly Openpath) is mobile-first and binds tightly to Avigilon video. That binding is the product, and it is excellent if you are buying the whole stack and a liability if you are not. LenelS2 and Genetec are the enterprise and upper-mid-market head ends, both running on Mercury hardware, which is why a building can change between them. Kantech’s single-door PoE and four-door IP controllers are a sensible mid-market multifamily answer where a full enterprise platform is not warranted.
Salto and standalone locks
Salto’s virtual network writes access rights onto the credential itself, so battery-powered locks on interior doors need no wiring at all. In a building where getting cable to a storage room, a roof door or a rear stair is genuinely impractical, that is a very large advantage. The trade is architectural and you must understand it: a revoked credential remains valid at an offline lock until the card next touches an updating point. Salto’s real-time and hybrid tiers address that, at cost. Alarm Lock Trilogy keypad locks are the budget rung of the same ladder: the entry-level unit holds around a hundred codes and keeps no audit trail at all, which means it can never answer “who opened this door”; the next model up adds one, and that single feature is usually worth the difference. Schlage covers the same territory with networked wireless and multifamily-specific lines.
Power supplies
Altronix is the field standard and this is the least glamorous, most consequential box in the room. What matters: individually fused outputs, integral battery charging, and a fire alarm input where the disconnect is selectable per output, because the maglocks on stair doors must drop on alarm and the fail-secure strikes on fire-rated doors must not. Getting that selection wrong is simultaneously a code violation and a security hole. The other number worth knowing is the low-voltage shutdown threshold, around 70 to 75% of nominal, which is why a building with a tired battery string sees doors dropping offline apparently at random long before the batteries are actually flat.
Locking hardware
HES, Von Duprin, SDC, Adams Rite, Alarm Controls and dormakaba for strikes, electrified levers and exit device hardware. The selection criteria that matter more than the brand: continuous duty where a door is held unlocked on a schedule, because intermittent-duty units burn out; DC rather than AC where silence is wanted, remembering that an AC strike buzzes by design and swapping to DC produces the “the buzzer does not buzz any more” callback in older buildings; a strike rated for the actual cycle count, since inexpensive units are rated around ten cycles a day and a lobby door can see hundreds; and a deadlatch finger that lands on the strike face rather than dropping into the pocket, or the door can be shimmed open by anyone with a credit card.
Reader protocol, which is a specification and not a brand
Wiegand sends the card number down two data conductors in the clear, tops out around 500 feet, needs a home run per reader, and cannot detect a device wired in behind the reader capturing every credential presented. OSDP runs over a twisted pair to roughly 4,000 feet, multi-drops several readers on one cable, supervises them so the panel knows when one goes offline or is swapped, and its secure channel encrypts the link with AES-128. On new work we specify OSDP. The step that is skipped constantly in the field is moving devices off the default installation key onto a unique secure channel base key. Without that, the encryption is theater.
Cameras, recorders and video management
Camera brand matters less than most buyers expect and sensor size, lens choice and pixel density on the target matter far more. A four-megapixel camera on a large sensor will beat an eight-megapixel camera on a small one in a dim lobby every time, because sensor area is what collects light.
Axis
The reference point. The best technical documentation in the industry by a distance, mature compression that cuts storage without visibly hurting the image, and a long track record on firmware support. Weaknesses: price, and a tendency to sell capability a storefront will never use. On a job where the image has to hold up in a dispute, it is money well spent; on a warehouse aisle it may not be.
Hanwha Vision
The most consistent sensor-to-price ratio in the market right now, with its own smart compression and a sensible product structure. A good default where the budget will not carry Axis across the whole site.
Bosch, Avigilon, i-PRO and Pelco
Bosch is conservative, durable and strong in difficult environments. Avigilon has excellent analytics that work best inside its own video management software. i-PRO and Pelco both remain credible where a specific optical or environmental requirement rules out the mainstream choices.
Verkada, Rhombus and Eagle Eye
Cloud-first video, and the appeal is real: no server to patch, no VPN to maintain, administration from a browser. The constraint is bandwidth and it is arithmetic, not opinion. Continuous cloud recording needs roughly one to two megabits per second of sustained upload per 1080p camera. Forty cameras is therefore around 80 Mbps sustained, around the clock, which a great many commercial circuits in this city cannot deliver without a fiber upgrade nobody budgeted for. The honest architecture in most New York buildings is hybrid: record locally, push events and selected streams to the cloud.
Digital Watchdog, Uniview and GeoVision
Value tiers that are perfectly serviceable when specified deliberately rather than because they were the cheapest line on a distributor’s page.
Milestone and Genetec
Video management software rather than cameras, and the right answer when a site is large enough that the recorder’s built-in interface stops being adequate, or when video and access control need to be operated as one system.
Storage, which decides whether any of this works
Retention is arithmetic and we show ours: one megabit per second of continuous video is 10.8 gigabytes per day. Eight four-megapixel cameras recording continuously at modern compression comes to roughly eight terabytes for thirty days before any redundancy overhead. Drives must be surveillance-rated (desktop drives are specified for something like eight to ten hours a day and typically fail inside six to eighteen months in a recorder), and with today’s very large drives, single-parity redundancy on a big array has a rebuild window long enough to be a genuine risk rather than a theoretical one.
Hikvision, Dahua and the rebranded lines built on them
Hikvision and Dahua make capable hardware at prices nobody else matches, and they are named on the FCC Covered List and restricted under NDAA FY2019 §889. The restriction binds federal agencies and their contractors, and it flows down to a great many grant-funded buyers: schools, housing authorities, transit, healthcare and nonprofits taking federal money. Under the second limb of §889, a federal contractor using covered equipment anywhere in its own operations is exposed, not just on the federal deliverable.
The trap is rebranding. Several consumer and value brands have historically been built on those platforms, and the box at receiving frequently says something different from the chipset inside. Verify the part, not the label.
What this means practically: if you are a private building with no federal nexus, your existing system is legal, it is not being recalled, and there is no rip-and-replace obligation. But new equipment authorizations were stopped in November 2022, so the parts and support pipeline is closing rather than growing. If federal money is anywhere near your project, we will not specify this equipment, and we will not make a blanket compliance claim about any product line without checking part numbers individually. A wrong answer there is a false claims problem, not a marketing one.
Cable, power and the passive layer
Here we specify by listing and standard rather than by brand, because on the passive layer the standard is the specification and the logo on the jacket is largely marketing.
- Fire listing follows the space, not the budget. Plenum-listed cable where the space handles environmental air, riser-listed cable in vertical shafts, general-purpose neither. Plenum may substitute downward for riser and general purpose; the reverse is never permitted and it is one of the most common reasons a riser pull fails inspection in this city.
- Never copper-clad aluminum. It is cheaper per foot, it fails on current handling under Power over Ethernet, and it is not listed for the application. If a bid is unusually cheap on cable, this is the first thing to ask about.
- Cat6 or Cat6A on new work, not Cat5e. The larger conductor has lower loop resistance, which matters at full length under a heavy PoE load where a meaningful fraction of the delivered power is lost in the cable itself as heat. The labor differential no longer justifies the saving.
- Certification testing, not a continuity beep. Testing runs to roughly five percent of installation cost and is what unlocks a manufacturer’s twenty-to-twenty-five-year system warranty. A contractor who does not offer it is asking you to take the cable plant on trust.
- Labeling to a scheme, at both ends. Along with an as-built and a labeling map handed over on paper. This is the single cheapest thing on a job and the first thing dropped when a schedule tightens.
Two active devices belong in this section because they solve problems nothing else does. Altronix power supplies and power controllers, covered above, for anything with a lock on it. And Telguard cellular communicators, which present a simulated dial tone to a panel that expects a copper phone line. That is the standard fix for a gate or entry panel stranded by copper retirement, and the same fix the building’s alarm dialers will need shortly afterwards.
The seven questions we run every product through
In this order. Brand preference is not on the list.
- Can the building physically carry it? Conductor count and condition in the riser, cable route length against the 100-meter channel limit, available power class at the device, and whether there is a pathway at all. This question eliminates more products than every other question combined.
- What is the five-year cost, including subscription? A cheaper panel with a permanent per-door or per-unit fee is frequently the more expensive purchase by year three. We put both numbers in the scope so the comparison is visible.
- What happens when the internet, the power or the vendor goes away? Documented offline behavior in writing: do cached credentials still work, for how long, can residents still be buzzed in, and what is the mechanical fallback. If a vendor will not answer that in writing, that is the answer.
- Is the credential actually encrypted, and who holds the keys? A 125 kHz proximity fob is cloneable in about fifteen seconds with a thirty-dollar device. Moving to 13.56 MHz is not automatically an upgrade: MIFARE Classic and the original iCLASS scheme are broken too. The chip, and whether the factory default keys were ever changed, is the security claim.
- Can somebody other than us service it? Open protocols where they exist, admin accounts in your name, documentation left behind. If the honest answer for a given product is no (and for some cloud intercoms it is), you get told that before you buy, not after.
- Is it listed for what it is being asked to do? UL 294 listing on access control equipment in an egress path, correct cable listing for the space, entrapment protection to UL 325 on a vehicle gate operator. Listing is checkable; adjectives are not.
- Does it create a federal nexus problem? If the building takes federal grant money, houses a federal contractor tenant, or expects to bid federal work, covered equipment is a liability regardless of how well it performs.
Common questions
If you are not a dealer, are you buying gray-market equipment?
No. Not being an authorized dealer is not the same as buying outside the legitimate supply chain. Equipment comes through normal authorized distribution, with manufacturer warranty intact and serial numbers that a manufacturer will recognize if you ever need to make a claim.
What we do not have is a contractual dealer relationship with any manufacturer, with the volume commitments and margin tiers that come with one. The practical difference to you is that no product on this page is being recommended because of a rebate threshold.
Where a manufacturer genuinely restricts something to authorized channels (certain extended warranties, certain enterprise licensing), we will say so at scope stage and, if it matters for your project, tell you to buy that item through an authorized dealer. It is a smaller list than the industry implies, but it is not empty.
For a typical pre-war walk-up with a dead intercom, what would you actually specify?
The honest answer is that the survey decides it, and specifically one measurement: how many conductors in the existing riser still ring out clean.
If the pair count and insulation are adequate, the strongest options are an analog bus platform such as the Aiphone GT or a two-wire video bus such as Comelit Simplebus 2, both of which ride the existing cable and turn a potential riser replacement into a few days of work. That path preserves the largest amount of your money.
If the riser is genuinely finished (too few usable conductors, or insulation that crumbles when disturbed), the real choice is between replacing the riser, which is expensive and permanent, and a cloud panel with no in-unit hardware at all, which deletes the riser problem and replaces it with a subscription for as long as the building exists. Neither is universally right. What is wrong is picking one before anyone has metered the cable.
We already have Hikvision cameras. Do we have to tear them out?
Almost certainly not, if you are a private building with no federal funding, no federal contractor tenant and no intention of bidding federal work. Existing legally installed systems can keep operating; there is no recall and no rip-and-replace mandate for a private commercial user.
What has changed is the support pipeline. New equipment authorizations for covered gear were prohibited in November 2022, retailers have delisted product, and the firmware and parts stream shrinks from here. So the practical planning question is not whether it is legal today but how long you want to depend on it.
The good news on cost: if you do replace, the cable, conduit, mounting points and often the PoE switching all stay. That is a large share of the original installation cost preserved, and it makes a phased camera-by-camera migration realistic instead of a single painful project.
Which do you push, cloud or on-premises?
Neither, and a contractor who has a universal answer to this is telling you about their business model rather than your building.
Cloud wins where administration frequency is high and IT capability is low: multifamily with constant tenant turnover, a small business with no server and nobody to patch one, portfolios where one person administers several sites. Directory changes take seconds in a browser instead of a truck roll, and the audit log survives a panel failure.
On-premises wins where the spend is capital rather than operational, where data residency matters, where the upload bandwidth for cloud video simply is not there, and where you want open-architecture panels you can keep while changing head-end software. The costs are a server, patching, backups and remote access to maintain.
What we insist on either way is that the outage behavior is documented in writing before purchase: what happens during an internet outage, a WiFi outage, a cellular outage and a power outage, and what the mechanical fallback is at every controlled door.
Can you service a system another contractor installed?
Usually, and it is a large part of what this work is. Diagnosis does not require a dealer badge. It requires knowing where the failure modes are, and those are properties of the architecture rather than of the brand.
The genuine limits are two. Some cloud platforms will not transfer administrative ownership without the original installer’s cooperation, which is precisely the lock-in this site keeps warning about; where that applies we will tell you what it will take to get control of your own system back. And some enterprise software requires a current licensed dealer to make configuration changes, in which case we can diagnose and do the physical work but the head-end change has to go through a licensed party.
For everything else (legacy amplifiers, strikes and door hardware, controllers, cabling, cameras and recorders), bring us the symptom and we will start with the event log, the voltage at the device and the door itself, in that order.
Other contractors display manufacturer logos. Does that mean anything?
Sometimes. A current, active dealer or authorization agreement is a real thing and a manufacturer will confirm it if you ask them directly, which is the test worth applying, because logos are easy to paste onto a web page and expired agreements rarely get removed.
What a dealer logo genuinely signals is training and support access on that product line, which has value. What it does not signal is that the product is right for your building, and it introduces exactly the commercial incentive this page exists to disclose.
The higher-value credentials to ask about are individual rather than corporate: specific manufacturer training a named technician has completed, and independent certifications from the industry bodies in this trade. Those attach to a person and are verifiable. Company-level logos frequently are not.
Work that usually comes with this
Intercom Systems
Installation, repair and replacement of entry intercoms, from a six-unit walk-up buzzer to a 200-unit IP video system.
Access Control
Fobs, cards, keypads and phone-based credentials, plus the locking hardware and egress compliance behind them.
Security Cameras
IP camera systems designed around identification distance and retention, not camera count.
Low Voltage & Cabling
The Cat6, fibre, risers, racks and wireless that everything else depends on.
Ask why, and ask why not the other one.
Tell us the building and the problem, and the recommendation you get back will name the alternative we rejected and the reason. That is the whole difference.
Sun to Thu 9am to 5pm · Fri 9am to 12pm · Sat closed